GLOBAL METHANE ASSESSMENT

Benefits and Costs of Mitigating Methane Emissions

Copyright © United Nations Environment Programme, 2021

This publication may be reproduced in whole or in part and in

any form for educational or non-profit purposes without special
permission from the copyright holder, provided acknowledgement
of the source is made. The United Nations Environment Programme
would appreciate receiving a copy of any publication that uses this
publication as a source.

No use of this publication may be made for resale or for any

other commercial purpose whatsoever without prior permission in
writing from the United Nations Environment Programme.

DISCLAIMER

The designations employed and the presentation of the material

in this publication do not imply the expression of any opinion
whatsoever on the part of the Secretariat of the United Nations
concerning the legal status of any country, territory, city or area
or of its authorities, or concerning delimitation of its frontiers or
boundaries. Moreover, the views expressed do not necessarily
represent the decision or the stated policy of the United Nations
Environment Programme, nor does citing of trade names or
commercial processes constitute endorsement.

Suggested citation: United Nations Environment Programme

and Climate and Clean Air Coalition (2021). Global Methane
Assessment: Benefits and Costs of Mitigating Methane Emissions.
Nairobi: United Nations Environment Programme.

ISBN: 978-92-807-3854-4

Job No: DTI/2352/PA

Global Methane Assessment / ACKNOWLEDGEMENTS 3

ACKNOWLEDGEMENTS
ASSESSMENT CHAIR
Drew Shindell

AUTHORS
A. R. Ravishankara, Johan C.I. Kuylenstierna, Eleni Michalopoulou, Lena Höglund-
Isaksson, Yuqiang Zhang, Karl Seltzer, Muye Ru, Rithik Castelino, Greg Faluvegi,
Vaishali Naik, Larry Horowitz, Jian He, Jean-Francois Lamarque, Kengo Sudo,
William J. Collins, Chris Malley, Mathijs Harmsen, Krista Stark, Jared Junkin, Gray
Li, Alex Glick, Nathan Borgford-Parnell

AFFILIATIONS
Duke University: Drew Shindell, Yuqiang Zhang, Karl Seltzer, Muye Ru, Rithik
Castelino, Krista Stark, Jared Junkin, Gray Li, Alex Glick
NASA Goddard Institute for Space Studies/Columbia University: Greg Faluvegi
Geophysical Fluid Dynamics Laboratory (GFDL): Vaishali Naik, Larry Horowitz,
Jian He
National Center for Atmospheric Research (NCAR): Jean-Francois Lamarque
Nagoya University: Kengo Sudo
University of Reading: William J. Collins
Stockholm Environment Institute (SEI): Johan Kuylenstierna, Chris Malley, Eleni
Michalopoulou
Colorado State University, Fort Collins: A. R. Ravishankara
International Institute for Applied Systems Analysis (IIASA): Lena Höglund-
Isaksson
Netherlands Environmental Assessment Agency (PBL): Mathijs Harmsen
Climate and Clean Air Coalition (CCAC): Nathan Borgford-Parnell

TECHNICAL REVIEWERS
Valentin Foltescu (Climate and Clean Air Coalition); Yangyang Xu (Texas A&M
University, USA); Durwood Zaelke (Institute for Governance & Sustainable
Development), Kristin Campbell (Institute for Governance & Sustainable
Development); Gabrielle Dreyfus (Institute for Governance & Sustainable
Development); Ben Poulter (National Aeronautics and Space Administration,
US); Kathleen Mar (Institute for Advanced Sustainability Studies); Ilse Aben
(Netherlands Institute for Space Research); Christopher Konek (United Nations
Environment Programme); Vigdis Vestreng (Norwegian Environment Agency);
Arif Goheer (Global Change Impact Studies Centre, Pakistan), Shaun Ragnauth
(United States Environmental Protection Agency).

Technical Advisor for the Global Change Assessment Model (GCAM):

Daniel Loughlin (United States Environmental Protection Agency)

COPY EDITOR: Bart Ullstein

MANAGING EDITOR: Tiy Chung

GRAPHIC DESIGN AND LAY-OUT: Katharine Mugridge

CONTENTS
FOREWORD 5
EXECUTIVE SUMMARY 7
The opportunity 9

Why act 11

How to do it 13
1. INTRODUCTION 17
1.1 Importance of methane 18
1.2 Purpose of this assessment 22
2. SOURCES OF METHANE 24
2.1 Anthropogenic sources 28
2.1.1 Bottom-up estimates 28
2.1.2 Regional differences in bottom-up estimates of emission sources 31
2.1.3 Estimating emissions using top-down methods from observed methane concentrations 33
2.2 Natural sources 37
3. COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 39
3.1 Experimental setup 42
3.2 Modelled ozone response to methane emissions changes 44
3.3 Modelled climate response to methane emissions changes 46
3.4 Human health impacts: methods and results 51
3.4.1 Ozone-related mortality 51
3.4.2 Heat-related mortality 57
3.4.3 Morbidity and labour productivity 60
3.5 Crop impacts: methods and results 68
3.6 Monetization of impacts: methods and results 74
4. METHANE EMISSIONS MITIGATION 86
4.1 Methane emissions under baseline and all-sector all-pollutant mitigation scenarios 89
4.2 Marginal abatement potentials and costs by sector 95
4.2.1 Broad sectoral analyses 95
4.2.2 Detailed subsector analyses 101
4.3 Abatement in response to methane-specific policies 108
4.3.1 Multiple Integrated assessment model 2050 analyses 108
4.3.2 2030 analysis using the Global Change Assessment Model 110
4.4 Methane mitigation and food 114
4.4.1 Reducing food waste 114
4.4.2 Improving livestock management 116
4.4.3 Dietary change 116
4.4.4 Conclusion on methane mitigation and food 117
4.5 Web-based decision support tool 118
5. CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES 119
6. REFERENCES 127
7. ABBREVIATIONS 149
8. ANNEX – ADDITIONAL FIGURES AND TABLES 154
Global Methane Assessment / FOREWORD 5

FOREWORD
Inger Andersen,
Under-Secretary-General of the United
Nations and Executive Director of the
UN Environment Programme

Today humanity is facing three systemic

and escalating planetary crises: the climate
crisis, the biodiversity and nature crisis,
and the pollution and waste crisis. These
crises are not independent but linked and
often stem from the same sources and
unsustainable models of consumption
and production. These links provide an opportunity to identify and
deploy solutions which deliver multiple-benefits and the ambitious
emissions reductions needed to overcome these crises.

This Global Methane Assessment highlights one of the greatest

opportunities available today to simultaneously address our
interlinked planetary crises and make peace with nature. Methane
is a powerful and short-lived climate pollutant which drives climate
change and harms human and ecosystem health by contributing to
the formation of ground-level ozone. Over the past decade, global
methane emissions have risen at a rate faster than at any time in the
last 30 years. While methane has both human and natural sources,
recent increases are attributed to activity in three anthropogenic
sectors, namely fossils fuels, waste and agriculture.

Current methane concentrations are well above those needed to

achieve our 2° C target. According to the 2018 Intergovernmental
Panel on Climate Change Global Warming of 1.5° C report, we
cannot achieve the Paris Agreement targets without immediately
reducing methane along with carbon dioxide and all other climate
forcing emissions.

Fortunately, there are readily available targeted control measures

that can reduce more than 30 per cent of projected anthropogenic
methane emissions this decade. Most of these technical solutions
are in the fossil fuels (oil, gas and coal) and waste sectors, and
Global Methane Assessment / FOREWORD 6

can be deployed at low or negative cost. These measures must

be implemented urgently and paired with increased efforts to find
breakthroughs in agriculture and nature-based solutions. Additional
measures that don’t specifically target methane, like shifting to
renewable energy, residential and commercial energy efficiency, and
a reduction in food loss and waste, can reduce methane emissions
by a further 15 per cent by 2030.

Readily achievable methane mitigation can deliver nearly 0.3° C of

avoided warming over the next two decades while simultaneously
reducing ground-level ozone concentrations. Such steps can avoid
255 000 premature deaths and help prevent more than half a million
emergency room visits from asthma every year. And because
ground level ozone also harms ecosystems and plants, the reduced
methane concentrations could increase global crop yields by 26
million tonnes per year.

Fast and ambitious methane mitigation is one of the best strategies

available today to deliver immediate and long-lasting multiple
benefits for climate, agriculture, human and ecosystem health.

Since its founding in 2012, the Climate and Clean Air Coalition has
led on global methane mitigation. As one of its founding partners,
the UN Environment Programme (UNEP) has helped lead the
Coalition’s pioneering oil and gas work through the Oil and Gas
Methane Partnership and the Global Methane Alliance. UNEP now
hosts the International Methane Emissions Observatory an initiative
that will integrate and reconcile methane emissions data, improve
reporting and verification of information, and validate monitoring
and measurement methodologies and techniques.

This year UNEP launched the UN Decade on Ecosystem Restoration.

The next ten years will be paramount in the fight to avert climate
change, the loss of nature and biodiversity, and the pollution and
waste crisis. Nations, businesses, international organizations and
individuals must commit to delivering methane mitigation consistent
with the conclusions of this report to make progress towards restoring
our degraded ecosystems and secure the future for humanity.
EXECUTIVE
SUMMARY
Global Methane Assessment / EXECUTIVE SUMMARY 8

R educing human-caused methane emissions is one of the most

cost-effective strategies to rapidly reduce the rate of warming
and contribute significantly to global efforts to limit temperature
rise to 1.5°C. Available targeted methane measures, together with
additional measures that contribute to priority development goals,
can simultaneously reduce human-caused methane emissions by as
much as 45 per cent, or 180 million tonnes a year (Mt/yr) by 2030.
This will avoid nearly 0.3°C of global warming by the 2040s and
complement all long-term climate change mitigation efforts. It would
also, each year, prevent 255 000 premature deaths, 775 000 asthma-
related hospital visits, 73 billion hours of lost labour from extreme
heat, and 26 million tonnes of crop losses globally (Figure ES1).

Figure ES1: Current and projected anthropogenic methane emissions and the
identified sectoral mitigation potential in 2030 along with several benefits
associated with sectoral-level methane emissions mitigation. Avoided warming
occurs in the 2040s, other impacts are annual values beginning in 2030 that would
continue thereafter.
Global Methane Assessment / EXECUTIVE SUMMARY 9

The findings in this assessment are the result of modelling that

uses five state-of-the art global composition-climate models to
evaluate changes in the Earth’s climate system and surface ozone
concentrations from reductions in methane emissions. Results allow
for rapid evaluation of impacts from methane emissions and the
benefits from mitigation strategies to the climate and ground-level
ozone formation and, air quality, public health, agricultural and other
development benefits. The assessment results are also available in a
web-based decision support tool1 that allows users to input different
methane emissions reduction goals to calculate the multiple benefits
at a national level.

THE OPPORTUNITY
•  ore than half of global methane emissions stem from human activities
M
in three sectors: fossil fuels (35 per cent of human-caused emissions),
waste (20 per cent) and agriculture (40 per cent). In the fossil fuel sector,
oil and gas extraction, processing and distribution account for 23 per cent,
and coal mining accounts for 12 per cent of emissions. In the waste sector,
landfills and wastewater make up about 20 per cent of global anthropogenic
emissions. In the agricultural sector, livestock emissions from manure and
enteric fermentation represent roughly 32 per cent, and rice cultivation 8 per
cent of global anthropogenic emissions. (Sections 1, 2.1 and 4.1)

•  urrently available measures could reduce emissions from these major

C
sectors by approximately 180 Mt/yr, or as much as 45 per cent, by 2030. This
is a cost-effective step required to achieve the United Nations Framework
Convention on Climate Change (UNFCCC) 1.5° C target. According to
scenarios analysed by the Intergovernmental Panel on Climate Change (IPCC),
global methane emissions must be reduced by between 40–45 per cent by 2030
to achieve least cost-pathways that limit global warming to 1.5° C this century,
alongside substantial simultaneous reductions of all climate forcers including
carbon dioxide and short-lived climate pollutants. (Section 4.1)

• here are readily available targeted measures that can reduce 2030
T
methane emissions by 30 per cent, around 120 Mt/yr. Nearly half of these
technologies are available to the fossil fuel sector in which it is relatively easy

1. http://shindellgroup.rc.duke.edu/apps/methane/
Global Methane Assessment / EXECUTIVE SUMMARY 10

to reduce methane at the point of emission and along production/transmission

lines. There are also available targeted solutions in the waste and agricultural
sectors. Current targeted solutions alone, however, are not enough to achieve
1.5o C consistent mitigation by 2030. To achieve that, additional measures must
be deployed, which could reduce 2030 methane emissions by another 15 per
cent, about 60 Mt/yr. (Sections 4.1 and 4.2)

•  oughly 60 per cent, around 75 Mt/yr, of available targeted measures have

R
low mitigation costs2, and just over 50 per cent of those have negative
costs – the measures pay for themselves quickly by saving money (Figure
SDM2). Low-cost abatement potentials range from 60–80 per cent of the total
for oil and gas, from 55–98 per cent for coal, and approximately 30–60 per cent
in the waste sector. The greatest potential for negative cost abatement is in the
oil and gas subsector where captured methane adds to revenue instead of being
released to the atmosphere. (Section 4.2)

•  he mitigation potential in different sectors varies between countries and

T
regions. The largest potential in Europe and India is in the waste sector; in China
from coal production followed by livestock; in Africa from livestock followed by
oil and gas; in the Asia-Pacific region, excluding China and India, it is coal and
waste; in the Middle East, North America and Russia/Former Soviet Union it
is from oil and gas; and in Latin America it is from the livestock subsector. A
majority of these major abatement potentials can be achieved at low cost, less
than US$ 600 per tonne of methane, especially in the waste sector and the coal
subsector in most regions and for the oil and gas subsector in North America.
(Section 4)

•  itigation potential from all measures is expected to increase between

M
2030 and 2050, especially in the fossil fuel and waste sectors. (Section 4.2)

•  he levels of methane mitigation needed to keep warming to 1.5°  C will

T
not be achieved by broader decarbonization strategies alone. The structural
changes that support a transformation to a zero-carbon society found in broader
strategies will only achieve about 30 per cent of the methane reductions needed
over the next 30 years. Focused strategies specifically targeting methane need
to be implemented to achieve sufficient methane mitigation. At the same time,
without relying on future massive-scale deployment of unproven carbon removal
technologies, expansion of natural gas infrastructure and usage is incompatible
with keeping warming to 1.5° C. (Sections 4.1, 4.2 and 4.3)

2. L
 ess than US$ 600 per tonne of methane reduced, which would correspond to ~US$ 21 per tonne of
carbon dioxide equivalent if converted using the IPCC Fifth Assessment Report’s GWP100 value of
28 that excludes carbon-cycle feedbacks.
Global Methane Assessment / EXECUTIVE SUMMARY 11

WHY ACT
•  ethane, a short-lived climate pollutant (SLCP) with an atmospheric
M
lifetime of roughly a decade, is a potent greenhouse gas tens of times
more powerful than carbon dioxide at warming the atmosphere. Methane’s
atmospheric concentration has more than doubled since pre-industrial times and
is second only to carbon dioxide in driving climate change during the industrial
era. (Section 1.1)
• Methane contributes to the formation of ground-level ozone, a dangerous
air pollutant. Ozone attributable to anthropogenic methane emissions causes
approximately half a million premature deaths per year globally and harms
ecosystems and crops by suppressing growth and diminishing production.
(Section 1.1 and 3.3)
• The atmospheric concentration of methane is increasing faster now than
at any time since the 1980s. In the absence of additional policies, methane
emissions are projected to continue rising through at least 2040. Current
concentrations are well above levels in the 2° C scenarios used in the IPCC’s
2013 Assessment. The Paris Agreement’s 1.5° C target cannot be achieved at
a reasonable cost without reducing methane emissions by 40–45 per cent by
2030. (Sections 1.1 and 4.1)
• The growing human-caused emissions come from all three sectors: fossil
fuels, agriculture and waste. (Section 1 and 4.1)
• Methane’s short atmospheric lifetime means taking action now can quickly
reduce atmospheric concentrations and result in similarly rapid reductions
in climate forcing and ozone pollution. Lower methane concentrations would
rapidly reduce the rate of warming, making methane mitigation one of the best
ways of limiting warming in this and subsequent decades. Doing so would also
help limit dangerous climate feedback loops, while simultaneously delivering
important health and economic benefits from reducing ground-level ozone.
(Sections 1.1 and 5)
• This assessment found that every million tonnes (Mt) of methane reduced:
- prevents approximately 1 430 annual premature deaths due to ozone
globally. Of those, 740 would have died from respiratory disease and 690
from cardiovascular disease. Every million tonnes of reduced methane
emissions could also avoid approximately 4 000 asthma-related accident
and emergency department visits and 90 hospitalizations per year.
(Section 3.4)
- avoids losses of 145 000 tonnes of wheat, soybeans, maize and rice
ozone exposure every year. This is roughly equivalent to increased global
yields of 55 000 tonnes of wheat, 17 000 tonnes of soybeans, 42 000
tonnes of maize, and 31 000 tonnes of rice annually for every million tonnes
of methane reduced. (Section 3.5)
- avoids the annual loss of roughly 400 million hours of work,
approximately 180 000 years, globally due to extreme heat3. Employment
within those sectors of the economy that are affected by heat exposure
varies between genders, leading to disparities in the impacts for men and
women that differ across countries. (Section 3.4)

3. L
 abour impacts of heat exposure are the most dependent on the size of the methane emission
changes, as well as background temperatures, and thus the linear scaling used here to obtain
impacts per million tonnes is only approximate.
Global Methane Assessment / EXECUTIVE SUMMARY 12

•  he global monetized benefits for all market and non-market impacts are
T
approximately US$ 4 300 per tonne of methane reduced4. When accounting
for these benefits nearly 85 per cent of the targeted measures have benefits that
outweigh the net costs. The benefits of the annually avoided premature deaths
alone from a 1.5°C-consistent-methane mitigation strategy is approximately
US$ 450 billion per year. (Sections 3.5 and 3.6)

• In addition to the benefits quantified here, methane reduction measures

also contribute to multiple Sustainable Development Goals (SDGs),
including climate action (SDG13), zero hunger (SDG2), good health and well-
being (SDG3). Additionally, they provide cost reductions and efficiency gains in
the private sector, create jobs, and stimulate technological innovation. (Sections
1.1, 4.4 and 5)

Figure ES2: Current and projected anthropogenic methane emissions and the
identified mitigation potential in 2030 of targeted controls and their costs (low cost
is greater than zero and less than 2018 US$ 600 per tonne of methane) as well as
a set of additional measures, such as energy efficiency and fuel switching, plus
behavioural changes, such as reduced food waste, dietary modification and energy
demand management, that are consistent with the 1.5° C long-term maximum
warming target. The cost of implementation of all-cost measures is more than 2018
US$ 600 per tonne of methane. Avoided warming is in 2040, other impacts are
annual values beginning in 2030 that would continue thereafter.

4. M
 onetary values are in 2018 US$. Fossil methane emissions are valued at ~$4 400 per tonne of
methane.
Global Methane Assessment / EXECUTIVE SUMMARY 13

HOW TO DO IT
Trends in methane emissions need to be reversed to achieve a multitude of
benefits by 2030. Targeted measures can be rapidly deployed to reduce methane
emissions from the fossil fuel and waste sectors, with a majority at negative or low
cost. To achieve targets consistent with keeping warming to 1.5° C, a combination
of targeted measures and additional measures which reduce methane but do
not primarily target it, are needed for all sectors, especially the agricultural one.
There are many potential strategies that would facilitate the implementation of
these measures including, for example, a price on emissions or an emissions
reduction target. There are also barriers. These include addressing the lack of
financing, enhancing awareness and improving education, changing production
methods, developing new policies and regulations, and changes in consumption
and consumer behaviour. Increased political will and private sector engagement
and action are needed.

The specific measures include the following.

•  il, gas and coal: the fossil fuel sector has the greatest potential for targeted
O
mitigation by 2030. Readily available targeted measures could reduce emissions
from the oil and gas sector by 29–57 Mt/yr and from the coal sector by 12–25
Mt/yr. Up to 80 per cent of oil and gas measures and up to 98 per cent of coal
measures could be implemented at negative or low cost. (Section 4.2)

• Waste: existing targeted measures could reduce methane emissions from the
waste sector by 29–36 Mt/yr by 2030. The greatest potential is in improved
treatment and disposal of solid waste. As much as 60 per cent of waste-sector
targeted measures have either negative or low cost. (Section 4.2)

• Agriculture: existing targeted measures could reduce methane emissions from

the agricultural sector by around 30 Mt/yr by 2030. Methane emissions from rice
cultivation could be reduced by 6–9 Mt/yr. The targeted mitigation potentials
from livestock are less consistent, ranging from 4–42 Mt/yr. Average cost
estimates vary across the available analyses. Behavioural change measures
and innovative policies are particularly important to prevent emissions from
agriculture, given the limited potential to address the sector’s methane emissions
through technological measures. Three behavioural changes, reducing food
waste and loss, improving livestock management, and the adoption of healthy
diets (vegetarian or with a lower meat and dairy content) could reduce methane
emissions by 65–80 Mt/yr over the next few decades. (Sections 4.2 and 4.4)
Global Methane Assessment / EXECUTIVE SUMMARY 14

•  dditional measures, which reduce methane emissions but do not primarily

A
target methane, could substantially contribute to methane mitigation
over the next few decades. Examples include decarbonization measures –
such as a transition to renewable energy and economy-wide energy efficiency
improvements. Various implementation levers exist. Emissions pricing, for
example, can be an effective policy which could incentivize substantial methane
mitigation and support the broad application of methane reduction measures.
A rising global tax on methane emissions starting at around US$ 800 per tonne
could, for instance, reduce methane emissions by as much as 75 per cent by
2050. (Section 4.3)

• Incomplete knowledge and monitoring of emissions in some sectors limits

the potential for technical mitigation innovation and strategic decision
making to efficiently reduce methane emissions. While there is enough
information to act now, addressing emissions at the scale and in the timeframe
necessary to meet the 1.5° C target will require an improved understanding of
methane emissions levels and sources. Continued and improved cooperation to
create transparent and independently verifiable emissions data and mitigation
analyses is needed. Such cooperative efforts would enable governments and
other stakeholders to develop and assess methane emissions management
policies and regulations, verify mitigation reporting and track emissions
reductions. (Section 5)

•  reater regional and global coordination and governance of methane

G
mitigation would support the achievement of the 2030 abatement levels
identified here. While methane reductions are increasingly being addressed
through local and national laws and under voluntary programmes, there are few
international political agreements with specific targets for methane. The Climate
and Clean Air Coalition (CCAC) leads global efforts to drive high-level ambition,
and strengthens national actions, polices, planning, and regulations around
methane mitigation. (Section 5)
Global Methane Assessment / EXECUTIVE SUMMARY 15

AN INTEGRATED APPROACH
Urgent steps must be taken to reduce methane emissions this decade. Given the
wide range of impacts from methane, the societal, economic, and environmental
benefits of acting are numerous and far outweigh the costs. The existence of
readily available, low-cost, targeted measures, and methane’s short-lived
atmospheric lifetime means significant climate and clean air benefits can be
achieved by 2030. Targets and performance indicators to reduce methane
must address the combined and multiple impacts methane has on climate, air
quality, public health, agricultural production and ecosystem health. Assessment
methodologies should be commonly agreed upon and transparent. An integrated
approach to climate and air quality mitigation is required to put the benefits of
methane reductions into context. To keep warming to 1.5° C, focused strategies
specifically targeting methane emissions need to be implemented alongside
substantial and simultaneous mitigation of all other climate pollutants including
carbon dioxide and short-lived climate pollutants.
Global Methane Assessment / EXECUTIVE SUMMARY 16

TARGETED MEASURES

Upstream and downstream leak detection and repair

Recovery and utilization of vented gas: capture of associated gas from oil wells;
blowdown capture; recovery and utilization of vented gas with vapor recovery units and
well plungers; Installation of flares.

FOSSIL FUEL Improved control of unintended fugitive emissions from the production of oil
SECTOR (OIL, and natural gas: regular inspections (and repair) of sites using instruments to detect
GAS, AND COAL) leaks and emissions due to improper operations; replace pressurized gas pumps and
controllers with electric or air systems; replace gas-powered pneumatic devices and
gasoline or diesel engines with electric motors; early replacement of devices with
lower-release versions; replace compressor seals or rods; cap unused wells.

Coal mine methane management: pre-mining degasification and recovery and

oxidation of ventilation air methane; flooding abandoned coal mines.

Solid waste management: (residential) source separation with recycling/reuse;

no landfill of organic waste; treatment with energy recovery or collection and flaring
of landfill gas; (industrial) recycling or treatment with energy recovery; no landfill
WASTE of organic waste.
SECTOR
Wastewater treatment: (residential) upgrade to secondary/tertiary anaerobic treatment
with biogas recovery and utilization; wastewater treatment plants instead of latrines
and disposal; (industrial) upgrade to two-stage treatment, i.e., anaerobic treatment with
biogas recovery followed by aerobic treatment.

Improve animal health and husbandry: reduce enteric fermentation in cattle, sheep
and other ruminants through: feed changes and supplements; selective breeding
to improve productivity and animal health/fertility

Livestock manure management: treatment in biogas digesters; decreased manure

AGRICULTURAL storage time; improve manure storage covering; improve housing systems and
SECTOR bedding; manure acidification.

Rice paddies: improved water management or alternate flooding/drainage

wetland rice; direct wet seeding; phosphogypsum and sulphate addition to inhibit
methanogenesis; composting rice straw; use of alternative hybrids species.

Agricultural crop residues: prevent burning of agricultural crop residues.

ADDITIONAL BENEFICIAL MEASURES

Renewables for power generation: use incentives to foster expanded use of wind,
solar, and hydro power for electricity generation.
FOSSIL FUEL
Improved energy efficiency and energy demand management: (residential) use
SECTOR (OIL,
incentives to improve the energy efficiency of household appliances, buildings, lighting,
GAS, AND COAL)
heating and cooling, encourage rooftop solar installations; (industrial) introduce
ambitious energy efficiency standards for industry; improve consumer awareness
of cleaner energy options.

WASTE Reduced consumer waste and improved waste separation and recycling, improved
SECTOR sustainable consumption.

Reduced food waste and loss: strengthen and expand food cold chains; consumer
AGRICULTURAL education campaigns; facilitate donation of unsold or excess food.
SECTOR
Adoption of healthier diets: decrease intake where consumption of ruminant products
is above recommended guidelines.

Table SDM1 Measures To Reduce Methane By 45 Per Cent By 2030

Note: The classification of measures as ‘targeted’ or ‘additional’ broadly reflects whether the measure’s
focus is on methane reductions or whether reductions occur largely as a co-benefit of measures with
another primary purpose, but final categorization is governed by the underlying literature on mitigation.
1. INTRODUCTION
CHAPTER FINDINGS
• M
 ethane amounts have risen rapidly over the
past decade, reaching five-year average growth
rates not seen since the 1980s.

• T
 he increase in atmospheric methane during
the 2010s has resulted in its amounts being well
above those in the 2º  C scenario used in the
Intergovernmental Panel on Climate Change’s
(IPCC) 2013 Assessment.

• T
 here is strong evidence that the increases
in methane amounts during the 2010s were
primarily attributable to increased emissions
from fossil fuel-related, agricultural and waste
sources, with fossil fuel-related activities
contributing as much as agricultural and waste
sources combined.

• Mitigation of methane is very likely the strategy

with the greatest potential to decrease warming
over the next 20 years.
 Global Methane Assessment / INTRODUCTION
1.1 IMPORTANCE OF METHANE
Methane (CH4) is a powerful greenhouse gas,
the atmospheric amount of which has more
than doubled since pre-industrial times (Nisbet
et al. 2019). It has been second only to carbon
dioxide (CO2) in driving climate change during
the industrial era (Myhre et al. 2013). Methane
is a short-lived climate pollutant (SLCP) with
an atmospheric lifetime of roughly a decade
(the perturbation lifetime, relevant for dealing
with emission reductions, is 12 years). Methane
contributes to the formation of tropospheric
ozone (O3), which, like methane, is a short-lived
but powerful greenhouse gas and surface ozone
is also an air pollutant with detrimental effects
on people, ecosystems and crops.
Emissions of methane to the atmosphere are
therefore harmful to society in multiple ways.
While methane is not directly dangerous to human
health, it does indirectly affect it and agricultural
productivity through ozone and climate change.
Recent studies have found evidence of these
consequences to health and agricultural damage
(Shindell et al. 2019) to be larger than previously
believed. These new studies include the finding
that tropospheric ozone may have much higher
impacts on public health, particularly respiratory
and cardiovascular deaths (Turner et al. 2016).
In addition, understanding of methane’s effect
on radiative forcing has recently improved,
leading to an upward revision since the Fifth
Intergovernmental Panel on Climate Change
1900
1850
GLOBAL MEAN AMOUNT (PPB)
1800
1750
1700
1650
•
1600
18
(IPCC) Assessment (Collins et al. 2018;
Etminan et al. 2016). Taken together, improved
understanding suggests that the overall societal
impact of methane emissions is likely larger than
indicated by prior estimates.
In large part, because of its impacts on public
health and agriculture, the broad social cost of
methane, that is the monetized societal damage,
including climate and air quality related impacts,
resulting from a tonne of emissions is 50–100
times greater, depending uon the preferred
discount rate (Shindell et al. 2017), than the
corresponding social cost of carbon dioxide,
before taking into account the recent updates
in knowledge.
The amounts of methane in the atmosphere have
changed dramatically over the past four decades
(Figure 1.1a). Amounts increased sharply in
the 1980s but slowed to a near constant level
between 2000 and 2005 when emissions and
sinks were roughly balanced (Figure 1.2).
However, atmospheric methane amounts have
increased rapidly again over the past decade
(Figure 1.1a). These observations demonstrate
that it is crucial to change the trajectory of this
greenhouse gas, which in 2020 exhibited the
highest growth rate in NOAA’s 37-year record
(Figure 1.1b). They also shows that atmospheric
methane responds quickly to reductions in
emissions, as shown by the negative growth
rates in the early 2000s (Figure 1.1b).
Figure 1.1a Global mean
methane amount,
1984–2019, parts per
billion
Source: Ed
Dlugokencky, NOAA/
ESRL (www.esrl.
noaa.gov/gmd/ccgg/
trends_ch4/)
 Global Methane Assessment / INTRODUCTION 19

15
ANNUAL METHANE GROWTH RATE (PPB/yr)

12

Figure 1.1b Global mean

6
methane annual growth
rate, 1984–2019, parts
3
per billion per year

Source: Ed
0 Dlugokencky, NOAA/
ESRL (www.esrl.
noaa.gov/gmd/
-3
ccgg/trends_ch4/
(accessed 6 February
•

-6
2021)
00

05

20
10
90

15
85

95

20
20
20

20
20
19

19
19

Figure 1.2 Estimated

EMISSIONS AND SINK (MILLION TONNES PER YEAR)

global methane
emissions and sinks
15 (removals), 1984–2019,
ANNUAL EMISSIONS million tonnes per year

12 Note: Estimated using

an inversion from
observed amounts
9
based on a constant
(time invariant) overall
6
global lifetime of
9.1 years and mass
balance (Section
3 2.1.3).
Source: Ed
0 Dlugokencky, NOAA/
ESRL
•

-3

-6
00

05

20
10
90

5
85

95

1
20
20
20

20
20
19

19
19

The possible reasons for these trends have been
discussed in a host of newly published papers
(Jackson et al. 2020; Saunois et al. 2020).
There is strong evidence that the increases in
the abundance of atmospheric methane are
due to growth in emissions rather than a slower
atmospheric removal rate (Figure 1.2). It appears
that increases are due to fossil fuel-related
activities and the combined contribution from
agricultural and waste sources in roughly equal
parts (Jackson et al. 2020).
Methane is emitted from both natural and
anthropogenic sources, as described more fully
in Chapter 2. Humanity has the greatest leverage
on anthropogenic sources, though human-
induced climate change and land-use changes
affect natural sources. Globally, anthropogenic
methane emissions are expected to continue to
increase. Under current policy scenarios, by 2030
anthropogenic methane emissions are expected
to increase by more than 15 per cent over 2010
levels, reaching nearly 380 million tonnes per
 Global Methane Assessment / INTRODUCTION 20

year (Mt/yr), an 8 per cent increase on 2020
levels (Höglund-Isaksson et al. 2020). Significant
emissions growth during the 2020s is expected
from all three sectors – fossil fuels, agriculture
and waste. The largest increases of about 13
Mt/yr are projected in the waste and wastewater
sectors, driven in particular by expected
strong growth in the population and income in
regions that currently have poorly developed
waste management systems. Emissions from
the fossil-fuel sector are projected to grow
by about 10 Mt/yr, with gas production and
distribution being the largest source, followed
by oil production whereas emissions related
to coal mining are projected to decrease very
slightly. The number of uncapped abandoned oil
and gas wells and coal mines is increasing and
could also continue to contribute to the growth
in emissions. In the agriculture sector, continued
growth in the number of ruminant animals, due
to rising demand for meat from an increasing and
increasingly affluent global population, will drive
increases in emissions from enteric fermentation
and manure management of around 6 Mt/yr.
Emissions from rice cultivation are, however,
likely to remain approximately constant.
For the 2015 United Nations (UN) Paris
Agreement to succeed, reducing anthropogenic
methane in addition to carbon dioxide is
paramount. Currently the largest contributor
to the departure from an idealized path to the
2º C target used in the IPCC’s Fifth Assessment
Report is the growth in methane amounts
(Figure 1.3). Achieving the more stringent
1.5º C target requires even larger decreases
in methane. The IPCC’s 2018 Special Report
concluded that reaching a sustainable mitigation
pathway to 1.5˚  C can only be achieved with
deep and simultaneous reductions of carbon
dioxide and all non-carbon dioxide climate
forcing emissions, including short-lived climate
pollutants such as methane. In 1.5˚ C pathways,
global anthropogenic methane emissions fall
by an average of 45 per cent by 2040, with the
full range showing decreases of 30–65 per cent
relative to 2010 by this time. Such trajectories,
which imply approximately a 2 per cent annual
decrease in emissions over the next 20 years,
are in sharp contrast to the current path which
shows an increase in the order of 0.5 per cent
per year and falls between methane abundance
in idealized paths leading to approximately
3–5º C of warming in 2100 (Nisbet et al. 2020).

80 CH4
Figure 1.3 Values for
70 the difference between
actual global mean annual
average radiative forcing
60
from major greenhouse
gases and values
50 under Representative
Concentration Pathway
DIFFERENCE (mW / m-2)

40 RCP2.6, 2000–2020,
milliWatts per square
30 metre

Note: Representative
20 Concentration
N 20 Pathway RCP2.6
10 (Collins et al., 2013)
as used in IPCC’s
0 Fifth Assessment and
C02 consistent with the
-10
Paris Agreement.

Source: Nisbet et
-20 al. (2020), updated
through 2020
-30 courtesy M. Manning
00

10

15

20
0

20
20

20
20
20

80 CH4

70
Global Methane Assessment / INTRODUCTION
The short lifetime of methane, and the quick
response of methane abundance to reduced
emissions described earlier, mean that any
action taken to reduce emissions will have an
immediate pay off for climate in addition to
the current and near-future human health and
agricultural production. Observations over the
past few decades have shown that decreased
emissions lead quickly to lower methane levels
relative to those that could be expected in the
absence of the decreases. That is, there are
no mechanisms that offset the decreases even
though there are significant natural sources.
Simply put, natural emissions do not make up
for the decrease in anthropogenic emission.
Indeed, the expectation that a reduction in
emissions will yield quick results, in the order
of a decade, is confirmed and emphasizes the
importance of methane.
Methane mitigation offer a way of rapidly
reducing the rate of near-term warming.
Also, mitigation of methane, along with non-
fossil greenhouse gases including some
hydroflurocarbons (HFCs) and black carbon-rich
sources of particulate matter (PM), is the only
plausible way of decreasing warming relative
to a reference case with minimal changes in
current policies over the next 20 years. This is
because a realistically paced phase-out of fossil
fuels, or even a rapid one under aggressive
decarbonization, is likely to have minimal net
impacts on near-term temperatures due to the
removal of co-emitted aerosols (Shindell and
Smith 2019). As methane is the most powerful
driver of climate change among the short-lived
substances (Myhre et al. 2013), mitigation of
methane emissions is very likely to be the most
powerful lever in reducing near-term warming.
This is consistent with other assessments; for
example, the Intergovernmental Panel on Climate
Change Fifth Assessment Report (IPCC AR5)
showed that methane controls implemented
between 2010 and 2030 would lead to a larger
reduction in 2040 warming than the difference
between RCPs 2.6, 4.5 and 6.0 scenarios.
(The noted IPCC AR5-era scenarios are called
representative concentration pathways (RCPs,
with the numerical value indicating the target
radiative forcing in 2100 (Kirtman et al. 2013)).
Though decarbonization is essential to meeting
long-term temperature goals, near-term
impacts are also important. Climate changes
over the next few decades will limit the ability
of human and natural systems to adapt. This
is especially problematic for the poorest and
women, both groups are particularly vulnerable.
21
Additionally, impacts such as sea-level rise and
glacier melting are influenced by cumulative
heat uptake, and many impacts may be non-
linear. Thus, there is a need for urgent action to
reduce damage from ongoing climate change. A
prior study (Shindell et al. 2017b) listed several
reasons for urgency, including stating that early
mitigation: “(i) reduces damage due to climate
change over the next few decades, including
those dependent on the pace of climate change
such as biodiversity losses; … (iii) reduces the
risk of potential non-linear changes such as
release of carbon from permafrost or ice sheet
collapse; (iv) increases the probability of staying
below 2° C through mid-century (Shindell et al.
2012; Ramanathan and Xu 2010); (v) reduces
long-term cumulative climate impacts; (vi)
reduces costs of meeting temperature targets
relative to late mitigation (Hamsen et al. 2016);
and (vii) stimulates progress toward the long-
term 2°  C target through achievement of near-
term benefits (Victor et al. 2016)”. Reductions in
methane emissions can thus provide near-term
climate benefits while carbon dioxide emission
reductions are implemented for long-term
stabilization.
Reducing methane emissions provides an
opportunity to rapidly change the environment,
both through surface ozone pollution and climate
change. Methane mitigation also has clear and
immediate economic benefits including improved
governance; cost reductions and efficiency in
the private sector; job creation, particularly in
relation to the leak detection and repair (LDAR)
within industry; and technological innovation and
entrepreneurship. Mitigating methane emissions
directly contributes to the achievement of
multiple Sustainable Development Goals (SDGs)
(Haines et al. 2017), including:
• S
 DG2 Zero hunger by improving ecosystem
health and agricultural yields, thereby helping
to end hunger and achieve food security;
• S
 DG3 Good Health and well-being, SDG11
Sustainable cities and communities and
SDG12 Responsible consumption and
production by reducing indoor and outdoor
air pollution, enabling sustainable food
production and helping ensure healthy lives
for people across the globe.
• S
 DG1 No poverty, SDG11 Sustainable cities
and communities and SGD13 Climate action
by helping reduce the exposure of vulnerable
populations to climate-related extreme events.
All 17 SDGs are shown in the Figure 1.4 below:
Global Methane Assessment / INTRODUCTION
Figure 1.4 The Sustainable Development Goals established by the United Nations in 2015
Source: United Nations
1.2 PURPOSE OF THIS ASSESSMENT
Despite the well-documented harms resulting
from its emissions, efforts thus far to reduce
methane have been inadequate given the rapid
growth in observed atmospheric methane
amounts. This is likely due in part to inadequate
awareness of the multiple benefits of emission
reductions. Additionally, a lack of adequate
policies, institutional structures and governance
play important roles. This assessment seeks
to better characterize the impacts of methane
reductions with state-of-the-art modelling,
and to document the multiple benefits that
could be realized through methane reductions
with an examination of near-term targeted
mitigation measures and additional measures
which contribute to other development priority
goals while also reducing methane emissions.
The classification of measures as ‘targeted’
or ‘additional’ broadly reflects whether the
measure’s focus is on methane reductions or
whether reductions occur largely as a co-benefit
of measures with another primary purpose such
as carbon dioxide reduction or health benefits.
Such a classification is subjective, however, and
the final categorization used in this assessment
is governed by achieving consistency with the
underlying literature on mitigation options. To
better quantify the benefits of methane mitigation,
and the uncertainties associated with such a
quantification, this work looks at the response
to emissions across a variety of impacts using
a range of models. The impacts were chosen
based on available evidence allowing specific
endpoints to be reliably quantified along with
an emphasis on links to the SDGs. Previous
analyses have shown the multiple benefits of
22
reducing methane emissions (Shindell et al. 2012;
UNEP/WMO 2011). This assessment adds to
those prior analyses by not only including more
modelling to better characterize robustness and
analyses of additional impacts, but also greatly
extends the scope of the associated economic
analysis.
In the troposphere, methane plays an important
role in regulating both the oxidation capacity as
a whole and the amount of ozone in particular.
Methane’s primary sink is oxidation through the
hydroxyl radical. When nitrogen oxides (NOx) and
sunlight are present, this oxidation process leads
to the production of ozone (O3). As ozone is also
a greenhouse gas, the net impact of methane
emissions on climate is larger than the direct
impact of the increased methane – ozone adds
about 38 per cent to the forcing associated with
methane alone (Myhre et al. 2013). In addition to
the warming caused by direct methane increases
and indirect tropospheric ozone increases, there
are smaller warming impacts from the oxidation
of methane. These include:
1. an increase in carbon dioxide in the
atmosphere;
2. an increase in water vapour in the
stratosphere; and
3. changes in atmospheric hydroxide (OH)
abundance that in turn affect concentrations
of radiatively active gases such as HFCs
and hydrochlorofluorocarbons.
These complexities mean that to obtain a
detailed understanding of the influence of
methane emissions requires the use of three-
Global Methane Assessment / INTRODUCTION 23

dimensional composition-climate models

capable of representing the most important of
these processes, as is done in this assessment.

The physical system modelling examines the

worldwide impact of methane emissions, which
have virtually the same impact regardless of
location (Fiore et al. 2008). This is because
although methane is a short-lived climate
pollutant, a class of pollutants with a lifetime
of roughly fifteen years or less, it is a relatively
longer-lived one and its atmospheric chemical
lifetime of nearly a decade is long compared
to atmospheric mixing timescales so that it is
a well-mixed gas, in the sense that geographic
variations in tropospheric concentrations are
relatively small. Impacts of methane emissions
include the responses of ozone, temperature
and precipitation, as well as subsequent
impacts of those changes on human health and
crop yields. Full three-dimensional atmospheric
composition-climate models are required to
capture the response of ozone, which is quite
inhomogeneous as it depends on the local
availability of other ozone precursors and
sunlight, as well as climate responses that are
also inhomogeneous.

This new modelling of the composition-climate

system is complemented by an analysis of the
potential for methane mitigation. This includes
both source specific analyses of the potential and
cost of targeted methane emissions abatement
as well as modelling with integrated assessment
models (IAMs) that incorporate such potentials
within their representation of the coupled
energy-economy-land system. This assessment
thus provides information on both the potential
for the abatement of methane emissions and
the impacts of such controls, both within this
document and in an online web-based decision-
support tool developed to support it. The
potential is compared with reductions in methane
emissions under scenarios consistent with
avoiding dangerous anthropogenic interference
in the climate system, the internationally agreed
goal of climate policy. This information can help
inform both large-scale efforts to set ambitious
goals for the mitigation of methane emissions
and detailed analyses of the most cost-effective
ways of mitigating emissions at global and
regional scales.
 2. SOURCES
OF METHANE
CHAPTER FINDINGS
• A
 nthropogenic emissions represent roughly 60 per cent of the total
methane emissions.

• A
 nthropogenic methane emissions come primarily from three sectors: fossil
fuels, ~35 per cent; agriculture, ~40 per cent; and waste, ~20 per cent.

• E
 missions from livestock are the largest source of agricultural emissions
with enteric fermentation the dominant process and cattle the dominant
animal causing the emissions.

• T
 he extraction, processing and distribution of the three main fossil fuels
lead to comparable emissions: gas and oil each contribute ~34 per cent,
followed by coal which contributes ~32 per cent of the fossil fuel methane
emissions.

• E
 missions vary greatly across regions, with the largest totals in China
and South Asia, but there are significant contributions from all regions.
Uncertainties in emissions are again largest for China but are also
significant for the Americas.

• A
 t subsectoral and regional levels, the largest emissions sources vary
markedly. For fossil fuels, they are coal mining in China, ~24 Mt/yr; oil
and gas extraction, processing and transport in West Asia, ~18 Mt/yr; in
Russia, ~15 Mt/yr; and North America, ~14 Mt/yr. Livestock emissions are
highest in Latin America, ~27 Mt/yr, followed by South Asia, ~22 Mt/yr.
Emissions from rice cultivation are highest in Southeast Asia+the Republic
of Korea+Japan, ~10 Mt/yr, closely followed by South Asia, ~8 Mt/yr, and
China, ~8 Mt/yr. Emissions from the waste sector are much more evenly
distributed around the world.

• A
 dvances in technology, especially remote sensing, are opening up new
opportunities to characterize emission sources more accurately, including
high emitters, and highlight substantial biases in much reporting at the
facility scale and even at a national scale.

• C
 hanges in emissions from natural sources of methane are likely to create
positive feedbacks as emissions increase in a warming climate.
Global Methane Assessment / SOURCES OF METHANE
Identifying and quantifying sources of methane is
a complex but critical element in understanding
atmospheric methane concentrations and
their trends. Methane has both natural and
anthropogenic sources that can be individually
small and geographically dispersed. While most
sources have been identified, their relative
contributions to atmospheric methane levels
remain uncertain. Estimates of global emissions
come broadly from two methods:
1. bottom-up which traditionally use some type
of activity data representative of the sector,
multiplied by an appropriate emission factor;
AND
2. top-down which use atmospheric
observations and models to indirectly infer
emissions.
Both bottom-up and top-down estimates are
discussed in this assessment.
The global methane budget is composed of
many sources and a small number of sinks
–atmospheric removal processes, primarily
oxidation by hydroxyl along with a small
contribution from uptake in soils (Table 2.1).
This table is based on bottom-up estimates as
these currently provide greater detail on specific
sources than top-down ones. The sources
and sinks are not currently in balance, and the
greater emissions flux relative to the sinks drives
the ongoing growth in atmospheric amounts.
The anthropogenic methane sources are the
focus of this assessment, but natural sources
are briefly described in Section 2.1.2 for
completeness. Note that this assessment
considers fires to be anthropogenic and wetland
emissions to be natural although both sources
are natural processes that are influenced by
human activities. This assessment does not
26
discuss the sinks in further detail as these are
relatively well constrained by observations of the
abundance of industrial trace gases whose only
loss mechanism is oxidation by hydroxyl, the
same loss process that dominates the methane
sink (Prather et al., 2012), and any trends in
methane sinks over recent decades are likely
to have been very small (Saunois et al. 2020).
It is noted, however, that there has been some
work on the potential for the active removal of
methane from the atmosphere in the future,
though such studies are at very early stages and
are their efficacy and economics remain to be
demonstrated (Jackson et al. 2020b; Lackner
2020; Jackson et al. 2019; Stolaroff et al. 2012).
It is important to emphasize that as the
atmospheric concentration of methane has
been measured to a high degree of precision
and the sinks and atmospheric residence time
are constrained, the total emissions are also well
constrained. This is because the atmospheric
abundance times, the removal rate, plus any
increase in concentrations, must balance the
total emissions flux, allowing the quantification
of emissions when the other factors are well-
known. Hence, although many individual source
strengths are not well known, any upward
revision in estimates of the magnitude of one
source would necessitate a downward revision in
another. Thus, the uncertainty surrounding total
emissions is substantially smaller than the sum
of uncertainties of individual emissions would
imply. Furthermore, isotopic data, in this case
the ratio of methane molecules to carbon atoms
containing seven neutrons and six protons (13C)
to the more common value of six neutrons and
six protons (12C)) constrain the apportionment
of emissions to fossil and biogenic sources,
so that the substantial uncertainties in many
individual natural biogenic sources are largely
independent from estimates of fluxes from fossil
fuel sources.
Global Methane Assessment / SOURCES OF METHANE 27

NATURAL MAGNITUDE ANTHROPOGENIC MAGNITUDE MAGNITUDE

SINKS
SOURCES (MT/YR) SOURCES (MT/YR) (MT/YR)

Wetlands 145 [100–183] Coal mining 44 [31–63] Soils 40 [37–47]

Total chemical
Termites 9 [3–15] Oil and gas industry 84 [72–97] 531 [502–540]
loss

Oceans 6 [4–10] Landfill and waste 68 [64–71] Total loss 571 [540–585]

Geological 45 [18–65] Ruminants 115 [110–121]

Wild animals 2 [1–3] Rice cultivation 30 [24–40]

Freshwaters 159 [117–212] Biomass burning 16 [11–24]

Permafrost soils 1 [0–1] Industry 3 [0–8]

Biofuels 13 [10–14]

Transport 4 [1–13]

Total natural 367 [243-489] Total anthropogenic 380 [359–407]

Total natural Total anthropogenic

232 [194-267] 364 [340-381]
(top-down) (top-down)

Table 2.1 Estimated natural and anthropogenic sources and sinks of methane, 2017

Note: Values shown are best estimates based on bottom-up studies (except last row), with the
minimum and maximum of the studies analyzed in that meta-analysis given in brackets. Natural
sources are based on 2000–2006 data as the separation into individual sources was not reported
for 2017. The total natural source was evaluated for 2017, however, and was identical to that for
the 2000–2006 period. The last row shows natural and anthropogenic totals based on top-down
constraints, highlighting how natural sources are substantially lower, based on top-down estimates,
likely due to overlap between wetlands and freshwaters, whereas anthropogenic bottom-up and
top-down estimates are more consistent.

Source: Jackson et al, (2020)

Global Methane Assessment / SOURCES OF METHANE
2.1 ANTHROPOGENIC SOURCES
Approximately 60 per cent of total global
methane emissions come from anthropogenic
sources. Of these, more than 90 per cent
originate from three sectors: fossil fuels, ~35 per
cent; agriculture, ~40 per cent; and waste, ~20
per cent.
Fossil fuels: release during oil and gas
extraction, pumping and transport of fossil
fuels accounts for roughly 23 per cent of all
anthropogenic emissions, with emissions from
coal mining contributing 12 per cent.
Waste: landfills and waste management
represents the next largest component making
up about 20 per cent of global anthropogenic
emissions.
Agriculture: emissions from enteric fermentation
and manure management represent roughly 32
per cent of global anthropogenic emissions.
Rice cultivation adds another 8 per cent to
anthropogenic emissions. Agricultural waste
burning contributes about 1 per cent or less.
These three categories are referred to as sectors
in this assessment, with further subdivisions
within the three termed subsectors. Biomass
burning, which has a mixture of anthropogenic
and natural causes, and biofuels are relatively
minor sources of methane emissions, as are
industry and transport. Emissions from some
subsectors show large uncertainties across the
existing inventories, in particular for coal mining
and rice cultivation (Table 2.1).
Source: Saunois et al. (2020). Emissions from EDGAR are the V4.3.2 version
Figure 2.1 Estimated anthropogenic methane emissions by sector from global inventories, 2017
28
2.1.1 BOTTOM-UP ESTIMATES
So called bottom-up emissions inventories are
created by compiling activity data and emission
factors for individual sectors, with the product
of these providing a total sectoral emissions
estimate. An example of these is rice cultivation
(hectare) for activity, and methane emissions per
hectare of rice cultivation for the emission factor.
These data are often estimated at the national
level. National statistics are frequently used for
activity estimates, whereas field and laboratory
measurements are generally used as sources of
data on emission factors.
Several groups of researchers have assembled
global emission inventories for methane
covering all sources, whereas other efforts have
focused on emissions from specific sectors.
A broad overview of available inventories for
anthropogenic methane emissions has been
assembled by the Global Carbon Project
(Saunois et al. 2020), incorporating data from
the Community Emissions Data System (CEDS)
(Hoesly et al. 2018), the Emission Database
for Global Atmospheric Research (EDGAR)
(Janssens-Maenhout et al. 2019), the United
States Environmental Protection Agency (US
EPA) (US EPA 2012), and the Greenhouse Gas
– Air Pollution Interactions and Synergies model
(GAINS) (Höglund-Isaksson 2020) as well as
several estimates for biomass burning emissions.
It shows that for all inventories, the two largest
sources are agriculture and waste together and
fossil fuels, respectively (Figure 2.1).
Global Methane Assessment / SOURCES OF METHANE
Within the fossil fuel sector, the oil and gas
subsector is the largest emitter, accounting for
roughly two-thirds of the sector’s total emissions,
with emissions from coal mining making up
most of the remainder. Emissions from industry
and transport are small, totalling ~7 Mt/yr, less
than 2 per cent of anthropogenic emissions,
from both categories (averaged across the four
inventories). Within the agriculture and waste
sectors, emissions related to livestock are the
largest source. These include emissions due to
enteric fermentation, primarily from cattle, and
emissions from manure. Landfills and waste
represent the next largest component, followed
by rice cultivation. In comparison, biomass
burning, which has a mixture of anthropogenic
and natural causes, and the use of biofuels are
relatively minor sources of methane. Agricultural
waste burning, included in the biofuels category
in the US EPA inventory and in the agricultural
sector in CEDS. GAINS, EDGAR and FAO
estimates for this category but not included in
Figure 2.1, range from 1 to 3 Mt/yr.
As noted, emissions for some subsectors
show large uncertainties across the existing
inventories, in particular for coal mining and
rice cultivation. Total anthropogenic emissions,
excluding biomass burning, are approximately
350–380 Mt/yr. For comparison, emissions from
wetlands are estimated to be roughly 160–210
Mt/yr based on biogeochemical models. Hence
approximately two-thirds of total emissions are
anthropogenic.
The two largest sources are livestock and fossil
fuels. Within the livestock subsector, enteric
fermentation and manure management are
the two processes generating emissions, with
the former dominant and cattle the dominant
animal (Figure 2.2). Within the manure category,
pigs play the largest role though cattle are
again important. Within the fossil fuel sector,
extraction, processing and distribution of the
three main fuels have comparable impacts, with
emissions from oil and gas each contributing 34
per cent followed by coal with 32 per cent of
sectoral emissions in 2020 (Höglund-Isaksson
2020). Emissions from the coal subsector are
entirely from mining-related activities, including
both active and abandoned facilities. Within oil
and gas, methane emissions associated with
onshore conventional extraction along with
downstream gas usage are the largest sources
29
(Figure 2.3). Venting, the deliberate release of
unwanted gas, is the primary cause of emissions
during onshore conventional extraction, whereas
fugitive emissions, the inadvertent release
or escape of gas from fossil fuel systems,
dominate downstream gas emissions. Within the
fossil fuel sector, at the national level, emissions
from the oil subsector in Russia and the coal
subsector in China appear to be far larger than
any other national level subsectors (Scarpelli et
al. 2020). While these types of data based on
national inventories are useful, it is important
to note that many local measurements show
large differences and often substantially higher
emissions than conventional reporting, in many
cases due to the presence of a small number of
super-emitters, and imply these estimates may
be too low (Zhang et al. 2020; Duren et al. 2019;
Varon et al., 2019; Zavala-Araiza et al. 2018).
These emissions give a sense of mitigation
opportunities by region and sector, which is
explored in Chapter 4.
Though some biomass burning is natural, current
burning results largely from anthropogenic
activities. Large amounts of biomass are burned
in the tropics in human induced fires related
to shifting cultivation, deforestation, burning
of agricultural wastes and the use of biofuels
(Dlugokencky and Houweling 2015). Biomass
burning remains a relatively small source of
methane and it accounts for approximately 5 per
cent of global methane emissions, an estimated
10–25 Mt/yr (Figure 2.1) (Saunois et al. 2020).
While rice cultivation feeds up to a third of the
world’s population, rice fields are a significant
source of methane (Mbow et al. 2019;
Dlugokencky and Houweling 2015). Methane
is produced through anaerobic decomposition
of organic material in flooded rice fields which
are responsible for approximately 8–11 per cent
of global anthropogenic methane emissions
(Saunois et al. 2020; Mbow et al. 2019).
As noted, there are substantial differences in
sector-specific bottom-up methane emissions
estimates created by different groups (Figure 2.1).
This difference indicates that these estimates
could be improved by additional work to better
quantify activity levels for each source, and
by improved quantification of source-specific
emission factors which may vary substantially
both between and within nations.
 Global Methane Assessment / SOURCES OF METHANE 30

CATTLE

BUFFALO

SHEEP

GOATS ENTERIC FEMENTATION

PIGS MANURE MANAGEMENT

CHICKEN

0 20 40 60 80 100 120

METHANE EMISSIONS (Mt)

Figure 2.2 Annual livestock methane emissions , million tonnes

Source: FAO (2013; 2017)

ONSHORE
CONVENTIONAL GAS

ONSHORE
CONVENTIONAL OIL

OFFSHORE OIL

OFFSHORE GAS

DOWNSTREAM GAS

UNCONVENTIONAL INCOMPLETE FLARE

GAS

VENTED
DOWNSTREAM OIL
FUGITIVE

UNCONVENTIONAL
OIL

0 5 10 15 20 25 30

METHANE EMISSIONS (Mt)

Figure 2.3 Annual oil and gas sector methane emissions by production type and reason, million tonnes
Source IEA (2020)

ONSHORE
CONVENTIONAL GAS
 Global Methane Assessment / SOURCES OF METHANE 31

2.1.2 REGIONAL DIFFERENCES IN BOTTOM-UP Asia+the Republic of Korea+Japan, ~25–35 Mt/

ESTIMATES OF EMISSION SOURCES yr; Africa, ~45–50 Mt/yr; Russia, ~20–25 Mt/yr;
Across multiple bottom-up inventories, West Asia, ~23–29 Mt/yr; and Europe, ~22–26
emissions are largest in China, ~46–74 Mt/yr, Mt/yr (Figure 2.4). Uncertainties about emissions
and South Asia, ~35–50 Mt/yr, but large totals from China are particularly large, but there are
are also emitted from North America, ~35–50 also substantial uncertainties about North and
Mt/yr; Latin America, ~50-60 Mt/yr; Southeast Latin America.

80

70
CEDS EDGAR

60
METHANE EMISSIONS (MT/YR)

GAINS USEPA
50

40

30

20

10

0
N

IL
CA

CA

IA

PE

IA

ST

IA

IA
IN

IC

IC
PA

AZ
AS

AS

SS

AN
EA
RO
RI

RI

ER

ER
CH
JA

BR

RU
AF

AF

TH

AL

CE
LE
EU
AM

AM
&

TR

D
U

O
N

ID
SO
&
ER

N
RE

M
RT

TI
AL

CE
H

LA
RT

O
RI

,K

N
O

F
O

IA

O
AT
N

AS

ST
U
EQ

RE
SE

Figure 2.4 Estimated anthropogenic methane emissions by region from global inventories, excluding biomass burning,
2017, million tonnes

Source: Saunois et al. (2020)

While all sources of methane emissions are important in every region, there are some significant
regional differences. Emissions from the coal and rice subsectors are particularly important in Asia.
Fossil fuel-related emissions are generally a large share of regional emissions throughout the northern
hemisphere. Solid waste separation and treatment offers opportunities across all regions. Figure 2.5
shows the global methane emissions for the year 2012 as reported by the EDGARv4.3.2 database,
with sector specific shares and total emissions for major world regions.
Global Methane Assessment / SOURCES OF METHANE
Figure 2.5 Global methane emissions for major world regions, 2012, million tonnes
Source: EDGARv4.3.2
The regional contribution to emissions from
individual sectors varies markedly. Within fossil
fuels, China is by far the largest source of coal-
mining related methane emissions at ~24 Mt/yr;
whereas for oil and gas the largest contributions
come from West Asia, ~18 Mt/yr; Russia, ~15
Mt/yr; and North America, ~14 Mt/yr (Figure 2.6).
Examining emissions related to livestock, the
largest contributions comes from Latin America,
~27 Mt/yr, roughly half of which is from Brazil;
and South Asia, ~22 Mt/yr; followed by nearly
equal contributions from China, Equatorial and
32
Southern Africa; Europe, North Africa and North
America, each with 8–11 Mt/yr.
Emissions from rice cultivation are dominated by
Southeast Asia+the Republic of Korea+Japan,
~10 Mt/yr; South Asia, ~8 Mt/yr; and China,
~8 Mt/yr, which together account for about 85
per cent of worldwide emissions. In contrast,
emissions from landfills and waste are much
more evenly distributed across the world.
These emissions give a sense of mitigation
opportunities by region and sector, which is
explored in Chapter 4.
 Global Methane Assessment / SOURCES OF METHANE 33

60
COAL MINING, 42 MT OIL AND GAS, 83 MT
7.9
RICE CULTIVATION, 30 MT ENTERIC FERMENTATION
50
AND MANURE, 110 MT
LANDFILLS AND
11.2 WASTE, 67 MT
METHANE EMISSIONS (MT)

40 8.3
0.4
11.4
30 9.8 0.6
0.2
10.4 2.3 3.0
1.2 21.7 0.1 0.1
13.6 5.2 2.0
0.6 7.4
20 0.2 3.6
8.0 4.9 11.0
9.0
6.8 24.0 5.9 14.6
5.2 14.1 13.4 0.1 18.1
10 3.3 8.2 6.6
0.6 2.5
3.6
6.3 6.0 7.9
2.0 2.8 2.9 4.8
4.1 2.5 3.0 0.6
0.2 1.7 0.6 0.3 2.0 1.0 0.1 2.7
0
IL

PE

ST
CA

CA

IA

IA

IA
IN

IC

IC
PA

AZ
AS

AS

SS
EA
RO
RI

RI

ER

ER
CH
JA

BR

RU
AF

AF

TH

AL

LE
EU
AM

AM
&

TR

D
U
N

ID
SO
&
ER

N
RE

M
RT

TI
AL

CE
H

LA
RT

O
RI

,K

N
O

F
O

IA

O
AT
N

AS

ST
U
EQ

RE
SE

Figure 2.6 Estimated annual sectoral methane emissions by region and global sector totals, excluding Oceania, 2017,
million tonnes

Source: Saunois et al. (2020).

2.1.3 ESTIMATING EMISSIONS USING TOP- Ground-based data is highly accurate and
DOWN METHODS FROM OBSERVED now fairly common. The first accurate in situ
METHANE AMOUNTS measurements of methane were made in 1978
(Blake et al. 1982). In 1983, measurement
To evaluate bottom-up inventories and obtain
stations from the National Oceanic and
a good quantitative understanding of methane
Atmospheric Administration (NOAA/ESRL)5
sources and sinks, it is critical to monitor and and Advanced Global Atmospheric Gases
quantify historic and current methane amounts. Experiment (AGAGE) (Prinn et al., 2000)6 had
This section gives a brief description of the global coverage. As shown in Figure 1.1a, these
different ways of measuring methane amounts networks show the amount of methane and
and producing emissions estimates, and Table its change over time. The latter is proportional
A4 in the Annex gives a summary of each to the imbalance between emission sources
measurement technique available as well as its and sinks. The globally averaged amount, for
advantages and disadvantages. Ground-based example, stabilized at approximately 1 770 parts
measurements, ice core data, observations per billion (ppb) around the year 2000, indicating
from aircraft, forward and inverse modelling and that at that time sources and sinks were equal
satellite data are all discussed. Results from (Figure 1.1).
such top-down methods generally compare
reasonably well with bottom-up estimates such To extend the record of methane amounts back
as those shown in Table 2.1 (Jackson et al. 2020; in time before ground-based measurements
Saunois et al. 2020). became available, several studies (Bock et

5. www.esrl.noaa.gov/gmd/ccgg/trends_ch4/ (accessed 18 February 2021)

6. https://agage.mit.edu/ (accessed 18 February 2021)
Global Methane Assessment / SOURCES OF METHANE
al. 2017; Chappellaz et al. 2000; Bender et al.
1997; Chappellaz et al. 1990; Stauffer et al.
1985) have been published on atmospheric
methane amounts reconstructed from polar ice
cores. These records provide further and strong
evidence of the predominant role of anthropogenic
sources of methane emissions in causing the
observed increases over the industrial era.
Another important contribution of the evidence
from ice cores is the range of emissions during
the Holocene7, which further demonstrates the
current rise is well outside natural variability.
These measurements significantly improve
the quantitative understanding of historic and
current changes in the global methane budget.
Ice core records published by Etheridge et al.
(1998), for example, showed that atmospheric
methane levels have almost tripled since 1800.
Understanding methane abundances in the
atmosphere can help constrain emissions from
particular sources. Several studies (Gasbarra et
al. 2019; Vaughn et al. 2018; Lavoie et al. 2017;
Robertson et al. 2017; Sweeney et al. 2015;
Karion et al. 2013) have used aircraft-based
measurements to estimate methane emissions
from individual facilities including landfills or
natural gas production regions, or specific regions
such as North America. One typical approach
includes transects at multiple altitudes around
a source while continuously measuring methane
concentrations, and wind speed and direction.
Figure 2.7 shows an example of a network that
contains measurements from aircraft.
As discussed in the National Academy of
Sciences report Methane Emission Measurement
and Monitoring Methods (2018), following in
situ sampling, the emission rates are estimated
using a mass balance approach in which the
concentration differences between the upwind
and downwind sections of the flight paths are
multiplied by the ventilation rate for the volume
enclosed by the flight paths to arrive at the
emission estimate (Conley et al. 2017; Gvakharia
et al. 2017).
7. the current geological epoch, which began approximately 11,650 years before present, after the last glacial period
34
While these techniques have been applied with
success to a variety of methane sources in urban
areas (Mehrotra et al. 2017; Cambaliza et al.
2015) they have limitations and restrictions that
need to be further investigated – this approach,
for example, can only detect emissions that
are encountered at flight elevations at the flight
radial distance.
Very recently, remote sensing from aircraft has
been used to quantify emissions from specific
sources with relatively high accuracy, based on
flights of independent instruments on different
aircraft (Gorchov Negron et al. 2020; Duren et
al. 2019; Baray et al. 2018; Johnson et al. 2017;
Frankenberg et al. 2016; Karion et al. 2013).
These data have shown that many bottom-
up estimates are incorrect. Measurements in
California, for example, revealed that multiple
landfills that had reported emission rates in the
range of 400–800 kilograms (kg) methane per
hour in fact had emission rates of ~1 200–2 200
kg methane per hour (Duren et al. 2019). Similarly,
measurements near offshore drilling platforms
in the Gulf of Mexico show that emission rates
were roughly double those in standard US EPA
inventories (Gorchov Negron et al. 2020). These
datasets also reveal that a few super-emitting
sources typically play an outsized role in the
total emissions from a specific source sub-
sector. Ground-based studies similarly show
that many bottom-up inventories underestimate
methane emissions relative to observations
(Chen et al. 2020; Alvarez et al. 2018; Brandt et
al. 2014). Though valuable, such ground-based
and aircraft data are currently available for only
a very small area as they require expensive field
and flight campaigns.
In addition to aircraft measurements, high altitude
platforms can be used to measure methane. High
altitude platforms are developmental vehicles
typically situated between 20 and 100 kilometres.
These platforms combine many of the advantages
of satellites and ground-based systems, providing a
flexible potential solution to many communications
challenges (Wang and Shao 2013).
Global Methane Assessment / SOURCES OF METHANE
Figure 2.7 Global Monitoring Laboratory network, 2020
Source: NOAA
To understand and quantify global and regional
budgets of methane and relate tropospheric
measurements to individual sources, forward and
inverse modelling can be used. Forward modelling
converts an emission source with a certain quantity
released per time into atmospheric abundance
in specific locations using bottom-up estimates
of emissions and sinks, as described in Section
2.1.1, along with an atmospheric transport model
to simulate atmospheric methane that can be
compared with observations.
Inverse modelling does the reverse where
measurements of atmospheric abundance are
used to calculate the location of an emission
source as well as the emission rate. An inversion
is a method that can be used to estimate
emissions of a gas at the surface of a domain
from atmospheric measurements of these gases
usually in mole fractions. Most inverse models
are based on Bayesian inference, in which
observations are combined with bottom-up
estimates (prior) resulting in estimated emissions
(posterior). By using inverse modelling to evaluate
bottom-up models of emissions, improvements
can be made to the bottom-up models.
Though most methane data through the end of
the 20th century comes from networks of ground-
based and airborne in-situ measurements, these
ways of measuring methane are generally labour
intensive and, in the case of most aircraft data,
35
because of their spatiotemporal scale, can
only provide snapshots of emissions. Another
limitation of these types of data is the limited
spatial coverage. There is great interest in and
calls for more satellite measurements of methane
to overcome some of these limitations.
Though still a relatively new data source, satellite
measurements of atmospheric methane have
already been used to detect emission hotspots
(Buchwitz et al. 2017; Kort et al. 2014; Marais et
al. 2014; Worden et al. 2012), to estimate emission
trends (Turner et al. 2016; Schneising et al. 2014)
and to derive emissions of large basins with oil
and gas extraction activities (Schneising et al.
2020; Zhang et al. 2020). They have been used
in global inverse analyses to estimate emissions
on regional scales (Miller et al. 2019; Maasakkers
et al. 2019; Alexe et al. 2015; Wecht et al. 2014;
Bergamaschi et al. 2013, 2009, 2007; Monteil et
al. 2013). In December 2019, satellite data were
used to quantify methane emissions from a gas
well blowout in Ohio, United States in February–
March 2018 (Pandey et al. 2019). That study,
using data from the Tropospheric Monitoring
Instrument (TROPOMI), demonstrated how strong
and effective satellite measurements can be in
detecting and quantifying methane emissions from
unpredictable events. Satellite data from GHGSat
and TROPOMI was also used to detect a major
continuous methane leak in central Asia which had
gone unnoticed (Varon et al. 2019).
Global Methane Assessment / SOURCES OF METHANE 36

Table 2.2 lists the types of instruments that have
been used to measure methane from satellites in
the recent past, along with those proposed for the
coming years. Satellite missions tend to be of two
types, global monitoring such as SCIAMACHY,
GOSAT and TROPOMI, or mission-focused ones
including GHGSat and MethaneSAT. The former
have very broad spatial coverage at relatively
low resolution whereas the latter are at higher
resolution but are targeted at specific locations
to monitor facility-scale amounts. With the rapid
advancements in the ability to monitor methane
emissions from space, several commercial
enterprises provide or intend to provide facility-
scale methane emissions information based on
satellite data, either using their own instruments,
such as GHGSat, Bluefield Technologies, or
combining data from government-launched
satellites with bottom-up information to better
identify sources, Kayrros, for example.
The overall understanding of methane emissions
by source is informed by both bottom-up
studies and top-down inversions based on
measurements. These provide complementary
constraints on the methane budget and are in
fairly good agreement for anthropogenic sources
though bottom-up estimates for natural sources
are generally much larger than top-down values
(Jackson et al., 2020; Saunois et al. 2020).

INSTRUMENT OR
AGENCY DATA PERIOD REFERENCE
SATELLITE

SCIAMACHY ESA 2003–2012 Frankenberg et al. (2006)

GOSAT JAXA 2009+ Kuze et al. (2016)

TROPOMI ESA, NSO 2017+ Butz et al. (2012)

GHGSat GHGSat, Inc. 2016+ Varon et al. (2018)

GOSAT-2 JAXA 2018+ Glumb et al. (2014)

geoCARB NASA (2022) Polonsky et al. (2014)

MethaneSAT EDF (2022) Benmergui (2019)

CO2M ESA (2025) Meijer et al. (2019)

MERLIN DLR and CNES (2021/2022)

Bluefield Bluefield Technologies (2023)

Sentinel-5 ESA, NSO (2023, 2030, 2037)

Table 2.2 Satellite instruments for measuring tropospheric methane

Source: Updated from Jacob et al. (2016). Dates in parentheses are intended launches.
Global Methane Assessment / SOURCES OF METHANE
2.2 NATURAL SOURCES
Natural sources of methane include gas hydrates,
freshwater bodies, oceans, termites and wetlands
as well as other sources such as wildfires. Values
for each of these are presented in Table 2.1.
Wetlands are defined as ecosystems in which
soils or peats are water saturated or where
surface inundation, whether permanent or
not, dominates the soil biogeochemistry and
determines the ecosystem species composition
(US EPA 2012). Examples of wetlands are bogs,
fens, marshes, muskegs, peatlands and swamps.
Globally, wetlands are the largest natural
source of methane with emissions, estimated
at 102–200 Mt/yr on average over 2008–2017
(Jackson et al. 2020; Saunois et al. 2020) which
is approximately one quarter of global methane
emissions. As can be seen from the range of
emissions estimates, however, there is significant
uncertainty surrounding methane emissions
from wetlands. Key to understanding methane
emissions from wetlands are three factors that
influence its production which are the spatial
and temporal extent of anoxia, linked to water
saturation; temperature; and substrate availability
(Saunois et al. 2020; Wania et al. 2010; Whalen
2005; Valentine et al. 1994). Wetlands are critical
to the current and future global methane budget,
but currently significant gaps remain in the
understanding of them (Sjögersten et al. 2020).
Freshwaters are another significant natural source
of methane, and in many cases these are similar
to wetlands and, as a result, double-counting is a
significant issue (Saunois et al. 2020).
Geological methane emissions from Earth’s
degassing are a major source to the atmosphere.
These emissions are from gas-oil seeps,
microseepage, mud volcanoes and submarine
seepage in sedimentary petroleum-bearing basins
and geothermal and volcanic manifestations
(Etiope and Schwietzke 2019). Bottom-up
estimates range from 27–63 Mt/yr, accounting
for ~8 per cent of total methane sources, and
most top-down estimates are consistent with
this range. More recently, however, using an
ice core record of the isotopic composition of
methane extending back to preindustrial times, it
was shown that geological sources were likely to
be much lower, in the range of 1–6 Mt/yr (Hmiel
et al. 2020). This suggests that this source may
have been considerably overestimated and that
anthropogenic fossil methane sources may be
underestimated by a similar amount, given the
relatively good constraints on the total emissions
of fossil methane.
37
Termites are a significant source of methane
and are responsible for approximately 1–3 per
cent of global methane – symbiotic micro-
organisms in their digestive parts of termites are
responsible for this. Methane emissions from
termites in studies conducted between 1982
and 2013 have, however, varied from 0.9 to 150
Mt/yr (Nauer et al. 2018) with the most recently
reported value being 9 Mt/yr (range [3–15] Mt/yr)
(Jackson et al. 2020; Saunois et al. 2020). This
significant uncertainty could be due to several
reasons, but the most prominent ones are:
a. lack of data on termite biomass required to
upscale (Nauer et al. 2018); and
b. a
 poor understanding of methane turnover
in the arboreal, epigeal or hypogeal nests
(mounds) of termite colonies (Kirschke et al.
2013; Brune et al. 2010; Bignell et al. 1997).
As discussed in Sjögersten et al. (2020),
research conducted in 2006–2019 by different
groups provides evidence that trees may be an
underestimated source of methane. Keppler et al.
(2006) suggested that plants produce methane
in aerobic conditions, and that the fluxes were
62–236 Mt/yr, or approximately 11–46 per cent
of the total global methane budget (Nisbet et
al., 2009; Keppler et al., 2006). It is argued that
the contribution of vegetation to the methane
budget is perhaps the least understood and
more recent estimates give a range of methane
emissions of 32–143 Mt/yr, approximately 22 per
cent of the total annual methane flux (Carmichael
et al., 2014). Significant uncertainties still remain
however, in particular, much of the flux from
trees occurs in wetland areas and so is already
accounted for in their estimated methane flux.
The ocean is a highly uncertain but small
contributor to the global methane budget, emitting
5–25 Mt/yr or 1–13 per cent of natural emissions
(Saunois et al. 2020; Weber et al. 2019). Possible
sources of oceanic methane include:
1. leaks from geological marine seepage;
2. production from sediments or thawing sub-
sea permafrost;
3. emissions from the destabilization of marine
hydrates, crystal water structures that trap
methane in sediments deposited on the
ocean floor;
4. in situ production in the water column,
especially in coastal areas because of
submarine groundwater discharge (Sharke
et al. 2014; Saunois et al. 2012).
Venting from the seabed usually does not release
methane to the atmosphere since it dissolves in
the ocean before reaching the surface. There are,
Global Methane Assessment / SOURCES OF METHANE
however, two well-known processes that emit
methane from the ocean into the atmosphere –
diffusive gas transfer and ebullition or bubbling
across the air–sea interface (Reeburgh 2007).
Further pathways have been identified that
may produce methane in the surface ocean
mixed layer, providing a more direct pathway for
methane to be emitted to the atmosphere (Weber
et al. 2019; Lenhart et al. 2016; Karl et al. 2008).
As noted above, several of the natural
methane sources potentially overlap with one
another, especially freshwaters and wetlands.
Presumably because of this, the total from
adding up each individual source based on
bottom-up estimates is in fact much larger
than the total value estimated using top-down
methods (Table 2.1). As such, this Assessment
relies upon the top-down constraint for the
total natural methane emissions, which leads
to relative source strengths for anthropogenic
and natural methane emissions of ~60 and 40
per cent, respectively. Taking into account the
recent findings of Hmiel et al. (2020) discussed
previously, the balance may be closer to two
thirds and one third, however.
Natural sources can also respond to climate
change, in which case a portion of the natural flux
38
is indirectly attributable to humans. Examples
include increased emissions from wetlands
under a warming climate (Shindell et al. 2013;
Gedney et al. 2004) and increased emissions
from thawing permafrost (Oh et al. 2020; Dean
et al. 2018; Cooper et al. 2017; Zhang et al.
2017; Schuur et al. 2015). Identified responses
of natural methane emissions to climate change
overwhelmingly represent positive feedbacks
and may be non-linear. An example is evidence
that the release of methane from thawing
permafrost occurs faster as a result of abrupt
rather than gradual thawing (Anthony et al. 2018).
These feedbacks cannot be mitigated directly
but can instead be reduced by the mitigation
of anthropogenic methane emissions and
emissions of other drivers of warming. It should
also be noted that increases in emissions due
to these feedbacks would increase methane’s
lifetime, so that the impact of each tonne of
anthropogenic methane would increase slightly.
Similarly, potential reductions in emissions of
nitrogen oxides, offset in part by reductions
in emissions of volatile organic compounds
(VOCs), would also increase methane’s lifetime.
In either case, the benefits of each tonne of
methane mitigation would be larger than those
analyzed here for current conditions.
 3. COMPOSITION-
CLIMATE
MODELLING AND
IMPACT ANALYSES
CHAPTER FINDINGS
• C
 oordinated modelling of both composition and climate responses to
methane changes was carried out for this assessment by five teams
using the models developed by the National Center for Atmospheric
Research (NCAR; CESM2; US); the Met Office and academia (UKESM1;
UK); the National Oceanic Atmospheric Administration Geophysical Fluid
Dynamics Laboratory (NOAA GFDL) (ESM4.1; US); the National Institute for
Environmental Studies (NIES), University of Tokyo and the Japan Agency
for Marine-Earth Science and Technology (JAMSTEC) (MIROC: Japan);
and the National Aeronautics and Space Administration (NASA) (GISS
E2.1; US). This provides improved characterization of responses of ozone
and climate and their uncertainties.

• B
 ased upon recently updated epidemiological results, the new modelling
indicates that there are 1 430 (670–2 110; 95 per cent confidence interval)
premature deaths due to ozone in response to each million tonne of
methane emitted, a value considerably higher than prior estimates.
Of those, 740 (460–990; 95 per cent confidence interval) are caused by
respiratory problems 690 (210–1 120; 95 per cent confidence interval) by
cardiovascular ones.

• R
 oughly 30 per cent of these premature deaths are of people aged less
than 70, and just more than 10 per cent of those aged less than 60.

• A
 t the national level, the largest total numbers of premature deaths are in
India, China, the US, Russia, and Japan, whereas the largest per person
impacts are in Ukraine, Lesotho, Egypt, Georgia and the Democratic
People’s Republic of Korea.

• T
 here are ~125 (60–185; 95 per cent confidence interval) premature
deaths due to heat exposure in China and the United States together per
million tonnes of methane emitted, and perhaps ~500 worldwide, though
confidence is low for the global values.

• T
 here are approximately 4  000 asthma-related accident and emergency
department (A&E) visits and 90 hospitalizations per million tonnes of
methane emitted.

• R
 oughly 300 (210–390; per cent confidence interval) million hours of work
are lost globally due to heat per million tonnes of methane emitted.

• 1
 45 (90–200) kilotonnes (kt) of wheat, soybeans, maize and rice are lost per
million tonnes of methane emitted. The largest impacts are in Brazil, China,
India and the US.

• M
 ortality impacts are valued at ~US$ 3 200 per tonne of methane using an
income elasticity of 1.0 across countries and ~US$ 6 700 using an elasticity
of 0.4, both undiscounted.
• Q
 uantified market costs are comparatively small at ~US$ 122/tonne and
stem predominantly from agricultural yield changes, and labour and
forestry losses.

• T
 he total valuation per tonne of methane is ~US$ 4 300 using a cross-nation
income elasticity of 1.0 and US$ 7 900 using an elasticity of 0.4. Valuation
is dominated by the effects of ozone, which are only weakly sensitive to
discounting, and thus the total valuation does not strongly depend upon
the relative weighting of impacts occurring at different times in the future.

• U
 ncertainties in health- or agriculture-related ozone metrics are roughly 25
per cent (95 per cent confidence interval) at the global level whereas for
surface temperature response they are ~55 per cent. Responses are nearly
linearly proportional to the methane change.

Modelling was undertaken using five state-of-the art global composition-

climate models that evaluate both the changes in surface ozone concentrations
and in the Earth’s climate system in response to methane reductions. This
coordinated comparison allows the investigation of the potential of reducing
warming and the impacts associated with ozone formation by decreasing
methane emissions, and the characterization of the robustness of these
responses.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
3.1 EXPERIMENTAL SETUP
The modelling focused on the response to 50 per
cent reductions in the anthropogenic increase
in methane amounts. A 50 per cent value was
chosen as it is large enough to give a clear signal
over meteorological noise in the models and
is similar in magnitude to what the worldwide
application of existing best targeted practices
could potentially achieve (Chapter 4). Based on
observations reported by NOAA’s Earth System
Research Laboratory, the 2015 mixing ratio of
methane was 1 834 parts per billion by volume
(ppbv). Ice core data indicate that the value in 1750
was about 722 ppbv, hence the anthropogenic
increase is 1 112 ppbv. A 50 per cent reduction
in that increase would therefore lower amounts
by 556 ppbv, so that the atmospheric abundance
would then be 1 278 ppbv.
The response is evaluated under present
day (2015) conditions and for two sets of
2050 emissions following a reference SSP3-
7.0 and short-lived climate pollutant control
scenarios developed previously by the Asia-
Pacific Integrated Model (AIM) group for other
pollutants that affect oxidation capacity (Table
3.1; simulations 1–6). The latter cases follow the
harmonized Shared Socio-economic Pathway 3
(SSP3) markerscenarios for the baseline (SSP3-
7.0) and short-lived climate pollutant controls
(SSP3-7.0_lowNTCF) (Fujimori et al. 2017).
These 2050 cases, with high and low emissions
of the non-methane ozone precursors nitrogen
oxides, carbon monoxide (CO) and non-methane
volatile organic compounds (NMVOCs) (Figure
3.1), are designed to provide information about
the sensitivity of the response to methane under
most conditions that are likely to be encountered
over the next few decades, potentially allowing
interpolation of methane responses to any given
set of emissions projections. Note that biomass
burning emissions were not changed in these
simulations.
Aerosol and aerosol precursor emissions
increase under the reference SSP3-7.0 case
in 2050, but these are unlikely to substantially
affect methane. One model performed additional
simulations with a 25 per cent decrease and 50
per cent increase in methane to test the linearity
of the response (Table 3.1; simulations 7 and 8).
Anthropogenic and biomass burning emissions
are from the SSP3 dataset produced in support
42
of CMIP6 and the IPCC AR6 cycle assessments.
For consistency with 2050 projections, 2015
emissions are taken from the same SSP dataset
which has been harmonized to 2015 historical
emissions reported by the Community Emissions
Data System (CEDS) project (v2017-05-08).
Five state-of-the-art composition-climate
models participated in the assessment: the
CESM2(WACCM6) model developed at the
National Center for Atmospheric Research
(NCAR) in Boulder, Colorado, US (Danabasoglu
et al. 2020; Gettelman et al. 2020); the GFDL
AM4.1/ESM4.1 (Horowitz et al. 2020; Dunne et
al. 2019) model developed by NOAA in Princeton,
New Jersey, US; the Goddard Institute of Space
Studies (GISS) E2.1/E2.1-G model developed by
the NASA in New York, New York, US (Gillet et
al. 2021; Kelley et al. 2020); the MIROC-CHASER
model developed jointly by the Atmosphere and
Ocean Research Institute, the University of Tokyo,
Kashiwa, the National Institute for Environmental
Studies (NIES), Tsukuba, the Japan Agency
for Marine-Earth Science and Technology,
Yokohama, and Nagoya University, Nagoya,
Japan (Sekiya et al., 2018; Watanabe et al.
2011; Sudo et al. 2002); and the UKESM1 model
developed jointly by the Met Office, Exeter, UK,
and the United Kingdom’s academic community
(Archibald et al. 2019; Sellar et al. 2019).
To reduce the burden imposed by running these
very computationally expensive models, only a
single pair of simulations were run with coupled
(i.e. responsive) ocean and sea-ice (denoted C1
and C2, with the C indicating coupled). Those
runs of 40–45 years length are used to assess
climate response to methane reductions. Other
runs examining the sensitivity of the ozone
response to methane changes were performed
in a configuration with fixed ocean boundary
conditions and, in most models, with fixed
meteorology and were therefore only run from
1–7 years. These shorter simulations include
a repeat of the base C1 and C2 simulations
in this configuration to assess the impacts of
the different experimental setup. Again, for
the sake of reduced computational expense,
the simulations were amount-driven, with
conversion to per million tonne emissions
performed afterwards (Section 4.1).
 Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 43

ANTHROPOGENIC EMISSIONS
METHANE
SIMULATION # MODELS CONFIGURATION
[PPB] CO, NOx, NMVOC BC, OC, NH3, SO2
[year] [year]
1 1 834
2015 2015
2 1 278
CESM2, GFDL
3 1 834 AM4.1,
SSP3-7.0
GISS E2.1, MI- Fixed sea surface
4 1 278 2050
ROC-CHASER, Temperatures/sea
UKESM1 SSP3-7.0
5 1 834 ice, fixed meteo-
SSP3-7.0_lowNTCF 2050
rology
6 1 278 2050

7 1 556
GISS E2.1 2015 2015
8 2 390
C1 1 834 CESM2, Coupled ocean and 2015 2015
GFDL ESM4.1, sea ice
C2 1 278 GISS E2.1-G 2015 2015

Table 3.1 Simulations performed for this assessment

SIMULATIONS 1,2,7,8 SIMULATIONS 3,4 SIMULATIONS 5,6

600 24 200

450 150
16
Mt/yr

300 100
8
150 50

0 0 0
CO BC OC NOX NMVOC NH3 SO2

Figure 3.1 Global anthropogenic emissions used in this analysis, million tonnes per year

Note: Simulations C1 and C2 used the emissions shown here for simulations 1 and 2, respectively.
Emissions for simulations 1, 2, 7 and 8 are from 2015; emissions for simulations 3 and 4 are from
SSP3-7.0 for 2050; and emissions for simulations 5 and 6 are from SSP3-7.0_lowNTCF for ozone
precursors and SSP3-7.0 for aerosols, both for 2050 (Table 3.1).

Source: UNEP and CCAC based on Fujimori et al. 2017.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
3.2 MODELLED OZONE RESPONSE TO
METHANE EMISSIONS CHANGES
Initially, the ozone response under present-day
(2015) atmospheric conditions (simulations 1
and 2) was examined. Reducing anthropogenic
concentrations by 50 per cent in this case led to
a reduction in the global area-weighted annual
average surface ozone of ~1.5–2 ppbv. Though
a small number, the reduction is larger in the
northern than the southern hemisphere and tends
to be greatest over land areas where methane
catalyzes ozone production more efficiently
than over the oceans owing to the presence
of anthropogenic nitrogen oxides. Hence the
global population-weighted exposure change is
~2–2.5 ppbv. More relevant to human health is
Figure 3.2 Change in annual average maximum daily 8-hour ozone exposure between the present day (2015) and
half anthropogenic methane simulations in various models and the multi-model mean (MMM)
Source: UNEP and CCAC
Exploring the sensitivity of the ozone response
to the background loading of other pollutants
(simulations 1–6), there is a clear dependence of
the response on the loading of nitrogen oxides.
Throughout the world, the ozone response
is larger when methane is altered within a
background atmosphere containing higher
levels of nitrogen oxides. The sensitivity varies
substantially from place to place depending on
other factors influencing ozone photochemistry,
including the availability of sunlight and of other
hydrocarbons or carbon monoxide that can
play the same role as methane. Despite the
complexity, it appears that this dependence can
44
the maximum daily 8-hour exposure averaged
over the year (MDA8), which is the metric most
closely linked to increases in premature deaths
from ozone in one of the largest epidemiological
studies to date (Turner et al. 2016). Changes in
this metric have been calculated for all individual
models as well as for the multi-model mean
(MMM) (Figure 3.2). Changes in this metric are
between 2–5 ppbv over much of the northern
hemisphere tropical and sub-tropical latitudes.
The GFDL AM4.1, GISS E2.1 and CESM2 models
are generally quite similar to one another, with
the UKESM1 model showing substantially larger
responses than those three and the MIROC-
CHASER model showing substantially weaker
responses.
be represented by a fairly simple logarithmic
equation that captures ozone sensitivity to
background nitrogen oxide conditions (Figure
3.3). Sensitivity to other pollutants is markedly
weaker. This indicates that the dependence of
ozone’s response to methane on background
nitrogen oxides can readily be captured in
policy-support tools to allow the estimation of
the impacts of changes in methane emissions
in any plausible near-future condition. This
should provide accurate results for fairly large
nations, for which domestic methane emissions
dominate nitrogen oxides, and of course requires
projected nitrogen oxides emissions levels.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 45

Figure 3.3 Country specific response of annual population-weighted average maximum daily 8-hour ozone exposure
due to a 50 per cent reduction in global methane concentrations, parts per billion

Note: The x axis is the country specific anthropogenic land emissions of nitrogen oxides from each
simulation pairing. The line is a logarithmic function for the multi-model mean, with the country
specific equation displayed in each panel.

Source: UNEP and CCAC

Next, the linearity of the response to different
magnitudes of methane concentration change
was examined. At the national level, population-
weighted ozone changes are extremely linear
across a range of methane increases and
decreases (Figure 3.4). Though the response
itself varies from country to country (i.e. the
slopes are different), the ozone change at the
national level is directly proportional to the
methane concentration change regardless of the
ozone metric chosen. This result is consistent
with prior studies which also indicate that the
ozone/methane relationship is approximately
linear (Fiore et al. 2008) but its magnitude
depends on the local availability of nitrogen
oxides, and, through nitrogen oxides, of hydroxyl
(West et al. 2006; Wang and Jacob 1998).

Figure 3.4 Relationship between changing global methane concentrations and the country-level population-weighted
ozone metric (parts per billion), with non-methane ozone precursor emissions (carbon dioxide non-methane volatile
organic compounds and nitrogen oxides held at 2015 levels (simulations 1, 2, 7 and 8 in Table 3.1)
Note: All results are from the GISS E2.1 model and are plotted as ozone changes relative to the metric
values from simulation 1 ([methane] = 1.834 parts per million). The selected countries are the 10 most
populous in the northern hemisphere.
Source: UNEP and CCAC
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
Given that the ozone response to methane
concentrations changes is quite linear with the
change in methane itself, the response of ozone
was converted into the response per unit of
methane emission change. For this analysis, the
response of methane concentrations per unit
methane emission change was first calculated,
using the relationship between methane
concentrations and emissions derived in prior
work (Fiore et al., 2008):
( )
f
C’ E’
— = —
C E
where C and C’ are initial and perturbed
concentrations, E and E’ are initial and perturbed
emissions, and f is the feedback factor defined
as the ratio of the perturbation lifetime to the total
atmospheric lifetime. As in Myhre et al. (2013),
f=1.34 was used and set total present-day
annual methane emissions at 568 Mt (Saunois et
al. 2016). Based on the simulations using values
of 1 834 and 1 278 for C and C’, E’ is found to
be 434 Mt/yr. Hence the concentration change
between simulations 1 and 2 corresponds to
emissions changes of 134 Mt/yr, so that the
methane response to emissions changes is 4.1
ppb CH4/Mt CH4. Note that using total present-
day methane emissions from another estimate
would modestly affect this result. For example,
using the 550 Mt/yr estimate of Kirschke et al.
(2013) would lead to emissions changes of 130
Mt/yr in simulations 1 versus 2.
The interannual variability of national and global
population-weighted annual MDA8 differences
were evaluated between runs 1 and 2 over three
years of simulation for the four models with
multiple years of simulations, except MIROC-
CHASER. At the global level, the standard
deviation based on interannual variability driven
by year-to-year differences in meteorology for
these ozone responses were 8.8, 5.3, 1.0 and
7.6 per cent respectively for CESM2, GFDL
AM4.1, GISS E2.1, and UKESM1. Values
were even smaller for MDA1. In comparison,
the 95 per cent confidence interval, based
on the variation across models in the global
population-weighted MDA8, was 24 per cent.
Values at the national scale showed a similar
range, except for some small countries, and
were still small in comparison with national level
variations between modes. Hence it appears
that variability within a model in the fixed Sea
Surface Temperature/meteorology setup is
relatively small, and hereafter only model-to-
model variability is incorporated along with
uncertainty in the exposure-response functions
46
in the uncertainty analysis as this provides a
good representation of the total. This analysis
also supports the inclusion of the MIROC-
CHASER single year simulations in this multi-
model assessment.
The effect of the model setup on the ozone
response was also examined by comparing the
changes in the coupled model simulations with
those in the simulations with fixed boundary
conditions – sea surface temperatures and sea
ice, along with meteorology in some models.
Again, examining the global population-weighted
annual MDA8, the differences between the 50
per cent anthropogenic methane reduction runs
including climate (C1 and C2) were found to
be very similar to those for the analogous runs
without a climate response (simulations 1 and
2). Specifically, responses in the coupled model
version were 3 per cent smaller for GFDL, 7 per
cent smaller for GISS, and 1 per cent larger for
CESM. As in the previous analysis, variability in
CESM is largest, but overall there was minimal
difference in the response over time in these
long simulations. For example, differences in
ozone responses over the last 10 years of those
simulations were typically within 3 per cent of
those in the first 10 years. Variations in response
due to setup, as with interannual variability, were
similar at the national scale except for some
small countries.
3.3 MODELLED CLIMATE RESPONSE
TO METHANE EMISSIONS CHANGES
Three of the five participating models were able
to complete the long climate simulations, C1
and C2 described in Section 3.1. The response
of these three were evaluated separately and for
the multi-model mean. Statistical significance
was assessed at the grid-point level by evaluating
the standard deviation across individual years
of model output – years 11–40 in each model’s
simulation. Temporal autocorrelation was
accounted for at each grid point by reducing the
sample size to represent an effective number
of independent data points, based on the lag-
1 autocorrelation (Zwiers and Von Storch 1995).
For the multi-model mean, this measure thus
includes both variations across the three models
and internal variability within the physical climate
system as represented by each individual model
and has been shown to give reliable statistical
ranges for sample sizes larger than about 50,
well below the sample size used of 90 (Conley
et al., 2018). For zonal and global mean
values, statistical significance was assessed
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
by examining the variability between models
in the long-term average surface temperature
response. Though atmospheric composition
will have largely equilibrated to the perturbed
methane within a decade in these experiments,
the climate system takes much longer to fully
respond. These results thus represent a partial
adjustment, as noted averaged over one to
three decades following the methane change.
To explore this topic further, the GISS E2.1
simulations C1 and C2 were extended by an
additional 40 years. Whereas the global mean
surface temperature response during years
11–40 was 0.20 ± 0.05° C, the response during
the next 40 years was 35 per cent larger with
a value of 0.27 ± 0.04° C (total response over
years 41–80). This suggests that the evaluation
of temperature responses provides a good
indication of the near-term impact of methane
changes, but that the longer-term changes
would be on the order of 35 per cent larger,
consistent with the slow response time of
the ocean. It is noted also that none of these
simulations include either the effect of carbon
dioxide produced from methane oxidation or
the response of the carbon cycle to methane-
induced warming. The former is estimated to
add ~2 per cent to the total radiative forcing from
methane (Myhre et al. 2013). The latter occurs
as the warming induced by methane emissions
suppresses carbon uptake, further enhancing
the temperature change attributable to methane
by ~10 per cent and extending its impact at long
timescales (Fu et al. 2020; Gasser et al. 2017).
The results for surface temperature show that
methane emissions, which lead to increases in
methane, tropospheric ozone and stratospheric
water vapour, cause clear and consistent zonal
mean, averaged over longitudes, warming over
most of the Earth during most seasons (Table
3.2 and Figure 3.5). Impacts are particularly
large over the mid-latitudes of the northern
hemisphere and especially over the Arctic.
The models were most consistent in the
response in the tropics. There is substantial
variation across the models in the annual mean
surface temperature response in the southern
hemisphere extratropics, however, especially
over the Southern Ocean, and even in parts
of the northern hemisphere – eastern Siberia/
Alaska and parts of eastern Europe – when
looking at the regional scale. It was noted that
the sample size was very small in this analysis
with just three models participating, and hence
the uncertainty in a quantity, such as the annual
global mean surface temperature change over
the 11–40-year period following emissions
47
changes, is likely smaller than a broader sampling
of models would yield for at least some globally
or zonally averaged results. In particular, the 95
per cent confidence interval in climate sensitivity
is estimated to be approximately ±34 per cent
(Sherwood et al., 2020), which translates into a 95
per cent confidence interval for the global mean
annual average surface temperature response to
a methane change of ±0.06° C (34 per cent of
0.18° C) rather than the ±0.02° C in this analysis.
It is recommended the larger uncertainty range,
~34 per cent relative uncertainty, is used in
impact calculations going forward.
Within the tropics, 30° S–30° N, the models give
very similar zonal mean responses (Table 3.2
and Figure 3.5). All suggest that warming over
the Pacific is less uniform than warming over the
Atlantic or especially the Indian Ocean basins
(Figure 3.6). Warming over tropical land areas
shows fairly good agreement across models
except over southernmost Africa. In contrast, the
consistently enhanced warming in the Arctic is
distributed differently in the three models. It thus
becomes difficult to precisely quantify climate
responses at national or even continental scales
over many parts of the world.
At seasonal timescales, the models are less
consistent over northern hemisphere mid-
latitude continental areas during boreal winter
than other seasons, as expected due to
the greater influence of dynamic relative to
thermodynamic response in that season (Table
3.2 and Figure 3.7). In the Arctic, however, the
models show both stronger and more consistent
responses during boreal winter/spring than
summer/autumn. Responses in the tropics and
southern hemisphere mid-latitudes are quite
similar across all seasons. In the Antarctic, there
is a high degree of variation across the models
in both austral autumn and winter seasons.
Seasonal values are often more variable at
smaller spatial scales but are nonetheless robust
for many large nations such as the Australia,
Brazil or the United States, or for some seasons
(Figure A1). There is little robust response in
precipitation at the local scale (Figure A2) or at
the zonal mean scale.
A large body of prior literature has shown that
the global mean surface temperature response
to well-mixed greenhouse gases scales fairly
linearly with the applied forcing (Collins et
al. 2013; Hansen et al., 2005). Although the
response to methane emissions is not exclusively
driven by the well-mixed gas methane, roughly
three quarters of the forcing comes from non-
ozone quasi-uniform forcing, due to methane,
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 48

stratospheric water and carbon dioxide; all
well-mixed greenhouse gases, and hence it
is assumed hereafter that it is a reasonable
approximation to scale climate impacts by the
methane emission change in a given case relative
to that applied in this analysis’ modelling.
Although changes in methane forcing are also
not perfectly linear with changes in emissions
or concentrations, these non-linearities are
relatively weak. Furthermore, the non-linearity of
forcing relative to concentrations tends to offset
the non-linearity in the concentration response
to emissions, so that the response of forcing to
emissions is closer to linear than either of those
relationships. A reduction of one quarter in the
concentration, for example, causes 48 per cent
as much forcing as reducing the concentration
by half rather than the 50 per cent one would
obtain with a linear change. Similarly, a reduction
in emissions of half the value used in simulations
1 versus 2 yields a concentration change, using
the equation given in Section 3.2, that is 51
per cent as large as that in the simulations,
again close to the linear value of 50 per cent.
Combining these, a reduction in emissions of
half the value used in the simulations hence
leads to almost exactly half the forcing used in
the simulations.

AREA ANNUAL DEC-FEB MAR-MAY JUN-AUG SEP-NOV

Global 0.18 (0.02) 0.19 (0.02) 0.18 (0.04) 0.17 (0.03) 0.17 (0.03)

60º N–90º N 0.28 (0.05) 0.56 (0.08) 0.31 (0.22) 0.07 (0.01) 0.19 (0.19)

30º N–60º N 0.22 (0.10) 0.19 (0.09) 0.19 (0.15) 0.25 (0.09) 0.26 (0.07)

30º S–30º N 0.16 (0.02) 0.16 (0.02) 0.16 (0.03) 0.16 (0.03) 0.18 (0.04)

60º S–30º S 0.12 (0.12) 0.13 (0.11) 0.12 (0.02) 0.11 (0.13) 0.12 (0.09)

90º S–60º S 0.19 (0.11) 0.20 (0.09) 0.29 (0.12) 0.24 (0.24) 0.05 (0.22)

Table 3.2. Multi-model mean area-weighted zonal mean temperature response (°C) to the methane increase from
one-half present methane to present value, along with the ozone and stratospheric water vapour responses to that
methane increase.

Note: values in parentheses are the 95 per cent confidence interval based on the model-to-model
variations, with results in bold statistically significant. These values are responses averaged over one
to three decades after emissions changes. Values over the subsequent four decades are ~35 per cent
larger and values including carbon cycles feedbacks are ~10 per cent larger, as discussed.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 49

Figure 3.5 Zonal mean annual average temperature

response to methane increases from one-half present
methane to present value, along with the ozone and
stratospheric water vapour responses to that methane
increase, degrees centigrade

Note: Shaded area indicated 95 per cent

confidence interval based on the three models.

Source: UNEP and CCAC

Figure 3.6 Annual average temperature response to methane increases from one-half present methane to present
value along with the ozone and stratospheric water vapor responses to that methane increase, degrees centigrade

Note: The multi-model mean (lower right) only includes values for areas in which the value is statistically
significant. Non-statistically significant areas are shown in grey.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 50

Figure 3.7 Seasonal zonal mean annual average temperature response to methane increases from one-half present
methane to present value, along with the ozone and stratospheric water vapor responses to that methane increase,
degrees centigrade

Note: Shaded area indicated 95 per cent confidence interval based on the three models.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
3.4 HUMAN HEALTH IMPACTS:
METHODS AND RESULTS
Both climate and ozone responses to methane
can affect human health through multiple
pathways. The relationship between ozone
exposure and respiratory mortality and morbidity
is well-established. Cardiovascular impacts
from ozone exposure have also been identified.
Climate change can affect human health in many
ways, including the spread of vector- and water-
borne diseases, undernutrition and food safety
and security-related illnesses, and exposure to
weather extremes such as heatwaves, floods and
storms. The focus here is on the current impacts
and projected risks of changing exposure to
high ambient temperatures as this is the best
quantified climate-related health impact.
In addition to the ozone and surface temperature
related impacts on human health quantified
here, other impacts from changes in methane
emissions are likely to exist but are not analyzed
in this assessment. Changes in stratospheric
ozone attributable to methane emissions, for
example, could affect human and ecosystem
health by altering the flux of ultraviolet (UV)
radiation reaching the surface. Policy changes
that affect methane emissions could also have
indirect effects on well-being. Fuel switching
away from natural gas, for example, could reduce
injuries related to pipelines. From January 2010
to November 2018, fossil fuel pipelines in the
United States are estimated to have resulted in
more than 5 500 accidents, with effects including
800 fires, 300 explosions, 600 injuries and 125
fatalities according to an analysis of data from
the Pipeline and Hazardous Materials Safety
Administration (Kelso 2018). Other potential
ways of reducing methane emissions, such as
transitioning to healthier diets in areas where
consumption of cattle-based products is above
recommended guidelines, could have profound
health benefits (Springmann et al. 2016). Hence
the effects reported here are likely to represent
a conservative estimate of the overall health
impacts of methane emissions or policy/
behavioural changes that could lead to methane
emissions reductions.
3.4.1 OZONE-RELATED MORTALITY
What are the impacts of ozone and heat exposure
on human mortality? Long-term exposure to
ozone can cause inflammation and allergic
responses leading to respiratory mortality, as
well as the development of a systemic oxidative,
proinflammatory environment that can increase
51
the risk of cardiovascular diseases. This analysis
of the impacts of ozone exposure on premature
death is based on the responses reported in
one of the largest studies to date that uses the
American Cancer Society Cancer Prevention
Study-II (ACS CPS-II) cohort (Turner et al.
2016). which calculated cause-specific deaths
attributable to incremental changes in MDA8
using a version of this cohort that spans 22 years
of follow-up and included 669 046 subjects of
whom 237 201 died. Here these results are used
globally, implicitly assuming that these long-
term ozone exposure-response relationships
are homogeneous across populations. While
there is evidence from other cohort studies in
Canada (Crouse et al. 2015) and in the United
States, including a cohort of ~61 million elderly
people (Di et al. 2017), showing a significant
relationship between long-term, seasonal to
annual, ozone exposure and premature mortality,
none of the smaller cohort studies conducted in
European countries have reported a statistically
significant relationship between ozone and
mortality (Bentayeb et al. 2015; Carey et al.
2013,). This may be due to differences in study
design, for example, the study of the French
cohort (Bentayeb et al. 2015) has a 95 per cent
confidence interval that is more than an order
of magnitude larger than other studies, such as
exposure estimation methods, length of follow-
up and number of events (Jerrett et al. 2013). No
studies are available for Africa, Asia, Australia or
South America.
Premature mortality attributable to long-term
ozone exposure is calculated using well-
established methods (Malley et al. 2017;
Anenberg et al. 2010), as described in the
equations below.
{
∆X = O
if [O3] ≤ TMREL
[O3] –TMREL if [O3] > TMREL
AF = 1 – exp -ß∆X
∆Mort = y0 × AF × Population
where TMREL is the theoretical minimum risk
exposure level (i.e. the counterfactual) in parts
per billion by volume, ΔX is the ozone exposure
in a particular grid box above the TMREL, β is the
exposure-response function – i.e. the slope of
the log-linear relationship between the change in
exposure and mortality from the epidemiological
study, AF is the attributable fraction of the
disease burden related to long-term ozone
exposure, y0 is the cause-specific baseline
mortality rate, population is the population
count in a particular grid box of people aged
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
30 years or older, and ΔMort is the estimated
number of premature, cause-specific mortalities
attributable to exposure to ozone levels above
TMREL. Note that TMREL is set simply as
the minimum ozone exposure reported in the
epidemiological study, in this case 26.7 ppbv
for the annual maximum daily 8-hour average.
There is no compelling evidence that exposure
to levels lower than those at which data is
available is safe, however. Hence this aspect of
the method is a conservative one.
Though the exposure-response function is
uniform worldwide, geographic variation in age-
binned values of both population and baseline
mortality rates is included. Population statistics
from the United Nations Population Division for
each country8 for the year 2015 are distributed
to grids using population density data from the
Gridded Population of the World (GPW) version
4 (CIESIN 2016). Baseline mortality rates, which
are the ratio of total deaths within a particular
age bin due to a particular cause, such as
respiratory deaths, to the total population in a
particular age bin, were derived by the Global
Burden of Disease (GBD) project (GBD 2017) for
each country of the world and the values for 2015
were used for consistency with the population
dataset. Global Burden of Disease baseline
mortality rates were mapped to best match the
current International Classification of Diseases
(ICD) codes for respiratory and cardiovascular
disease related deaths (respectively ICD-10
Codes: J00-J98; GBD Codes: B.3, A.2.3, A.2.4,
and ICD-10 Codes: I20-I25, I30-I51, I60-I69, I70;
GBD Codes: B.2.2, B.2.3, B.2.8, B.2.9, B.2.10)
for which significant impacts were found in the
epidemiological study (Turner et al., 2016). In
addition to this primary analysis based upon
2015 population and baseline mortality, an
analysis focused on the year 2030 was also
performed to explore how the benefits of future
policies might compare to those from changes
in the recent past or present. Projections for
2030 baseline mortality rates are obtained from
the International Futures project version 7.53
(International Futures 2020) for respiratory and
cardiovascular deaths by age group. Population
projections for 2030 were taken from SSP2
though these are fairly similar across the shared
socioeconomic pathways at this date (KC and
Lutz 2017). The geographic distribution of
population is unchanged in the 2030 analysis,
so too is the age distribution of the population
– only population totals are projected under the
shared socioeconomic pathways.
8. https://esa.un.org/unpd/wpp/DataQuery
52
Note that a statistically significant impact on
deaths associated with diabetes was also
reported but resulted in few total deaths and is
not included here. Cause-specific risk increases
per 10 ppbv were 1.12 (1.08–1.16; 95 per cent
confidence interval) for respiratory diseases and
1.03 (1.01–1.05; 95 per cent confidence interval)
for cardiovascular diseases based on the two-
pollutant model, accounting for exposure to fine
particulate matter, of the epidemiological study
(Turner et al. 2016). All impacts reported here are
based on ozone responses to sustained (rather
than pulse) methane emission changes.
Overall, in the primary analysis for 2015
conditions, the 50 per cent reduction in methane
leads to 99 000 (62 000–132 000; 95 per cent
confidence interval) fewer premature respiratory
and 92 000 (28 000–150 000; 95 per cent
confidence interval) fewer cardiovascular deaths
due to ozone exposure, which corresponds to
~1 400 fewer total deaths per million tonnes
of methane emission reductions. Of these, a
similar fraction is attributable to respiratory and
cardiovascular related deaths, respectively 740
(460–990; 95 per cent confidence interval), and
690 (210–1120; 95 per cent confidence interval),
with the uncertainty range on the latter markedly
larger. At the national scale, this leads to the
largest total impacts occurring in countries with
large populations, namely China, India, Japan,
Russia and the United States (Figure 3.8). On a
per person basis, however, impacts maximize in
lower latitude regions where ozone responses
are largest and especially in those nations where
baseline health conditions are relatively poor.
Per person impacts are greatest in Ukraine,
Lesotho, Egypt, Georgia, the Democratic
People’s Republic of Korea, Afghanistan,
Bulgaria, Serbia, Uzbekistan and Eritrea (in that
order). The United Kingdom, at 13th, is the highest
impacted on a per person basis of Organisation
for Economic Co-operation and Development
(OECD) countries.
In the analysis using 2030 population and
baseline mortality, the 50 per cent reduction in
methane leads to 105 000 (66 000–140,000; 95
per cent confidence interval) fewer premature
respiratory deaths and 81 000 (25 000–
132 000; 95 per cent confidence interval) fewer
cardiovascular deaths due to ozone exposure.
Respiratory-related deaths hence increase
by 6 000 relative to the 2015 analysis, due to
increases in, for example, India, 1400; the
United States, 600; Indonesia, 300; Mexico,
200, and many countries in Western Europe
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
that are partially offset by decreases in China,
-600, and many countries in Eastern Europe. In
contrast, cardiovascular-related deaths for 2030
are 11 000 fewer than those in the 2015 analysis
due to decreases in, for example, China, -6 000;
Russia -1000; Germany, -400; India, -500;
Indonesia, -400; Japan, -400; and the United
States, -400, that are offset by increases in some
other nations such as Pakistan, +200. There is
a divergence in response between respiratory
and cardiovascular deaths due to differences
in projected changes in baseline mortality
rates. In China, for example, mortality rates for
cardiovascular disease are projected to decrease
across all age groups whereas mortality rates for
respiratory disease are projected to drop only
for those aged 45–70 and rise for both younger
and older populations. In the United States,
mortality rates for cardiovascular disease are
again projected to decrease across all age
groups whereas mortality rates for respiratory
disease are projected to drop for those aged
less than 80 while rising for the 80+ age group,
with a decrease in mortality rates in some
younger populations, such as those aged 75–
80, being outpaced by the projected growth in
the population within that age bin. Similar trends
are at work in other nations, for example, in India
population is projected to increase by 17 per
cent but this is outweighed for cardiovascular
deaths by projected reductions in baseline
cardiovascular mortality rates which is not the
case for respiratory health. Given the generally
modest differences between these results, only
the primary 2015 analysis is reported hereafter
but it should be noted that, in particular for
analyses focused only on the better-constrained
respiratory-related deaths, the 2030 global
impacts are likely to be ~6 per cent larger.
Globally, roughly 40 per cent of avoided
respiratory-related premature deaths are in those
aged 80+, with about 28 per cent for people in
their 70s and 18 per cent for those in their 60s.
These values vary substantially, however, at the
national level. Two thirds to three quarters of
avoided deaths are for 80+ year-olds in many
European countries, including France, Sweden,
Switzerland and the United Kingdom, whereas
that age group experiences less than one third of
the total impact in several populous developing
countries such as India, Indonesia and Pakistan,
which instead see larger effects on populations
53
in their 70s and sometimes in their 60s. Values
are similarly distributed for cardiovascular
deaths. At the global level, for example, 36
per cent of avoided premature cardiovascular-
related deaths are people aged 80+, 36 per cent
for those in their 70s and 20 per cent for people
in their 60s. Unlike age, current epidemiological
studies do not report whether there are gender
differences in exposure-response functions.
As shown in Figure 3.4, the ozone exposure
metrics respond nearly linearly to methane
changes. Using the methods described to
calculate premature mortality attributable to
long-term ozone exposure, the linearity of this
impact to changes in methane were investigated.
It was assumed, based on Figure 3.4, that the
MDA8 ozone responds linearly to methane
changes and so this test examines the linearity
of the exposure-response function. Over a wide
range of methane emission reductions, national-
level responses are highly linear (Figure 3.9). This
is because the national average MDA8 ozone
changes in response to methane are fairly small,
ranging from a 1 to 4 ppbv decrease in response
to half anthropogenic emissions reductions
–134 Mt. Such small changes in methane do
not induce substantial non-linearity through the
exposure-response exponential function, nor
do they cause national exposures to pass the
TMREL threshold. The threshold is not passed
as countries in the Northern Hemisphere are
typically well above the 26.3 ppbv threshold
used here, whereas the few nations with very
low exposures, such as New Zealand or Papua
New Guinea are already below the threshold.
A few countries, including Australia, Colombia
and Uruguay, have national exposures within a
few parts per billion by volume of the threshold
and hence could easily see non-linear behaviour
were methane to be reduced in concert with
other pollution changes, but with methane
changes alone only a few island countries
cross the TMREL threshold. These results
hold for both respiratory and cardiovascular
premature deaths. It can thus be concluded that
the modelled response to a 134 Mt change in
methane emissions can be linearly scaled to
yield a highly accurate estimate of per million
tonne responses. Scaled results for the countries
most affected are given in Table 3.3 and a full list
is given in the Annex.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 54

Figure 3.8 National total premature respiratory death changes between present-day (2015) and one-half
anthropogenic methane simulations in five different models and the multi-model mean

Note: Values are given as totals and per 100 000 people aged 30 or over.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 55

Figure 3.9 National-level reductions in ozone-related premature respiratory deaths associated with the given
decrease in methane emissions, 5 selected countries

Note: Developed using the exposure-response function described in the text and assuming a linear
response of the maximum daily eight-hour average ozone to methane change.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 56

Table 3.3 Modelled total deaths and per million people aged 30+ per 10 Mt methane, 10 highest countries

RESPIRATORY

PER
COUNTRY TOTAL LOW HIGH COUNTRY LOW HIGH
MILLION
India 2 045 1 169 2 863 Lesotho 5.1 3.5 6.5
China 1 419 948 1 831 Korea Dem. People’s Rep. 4.5 3.2 5.7
United States 428 287 553 Eritrea 4.1 2.5 5.6
Japan 251 176 316 United Kingdom 3.8 2.5 5.0
Brazil 160 100 216 India 3.5 2.0 4.9
Pakistan 159 98 215 Portugal 3.2 2.1 4.2
United Kingdom 157 102 207 São Tomé and Principe 3.2 2.0 4.3
Indonesia 154 86 219 Guinea-Bissau 3.1 1.6 4.6
Germany 121 79 160 Zimbabwe 3.1 2.0 4.1
Russia 109 71 143 Denmark 3.1 2.1 4.0

CARDIOVASCULAR

PER
COUNTRY TOTAL LOW HIGH COUNTRY LOW HIGH
MILLION
China 1 566 502 2 522 Ukraine 6.5 2.1 10.8
India 1 122 297 1 844 Georgia 5.4 1.8 8.7
Russia 379 122 624 Uzbekistan 5.3 1.8 8.4
United States 333 108 540 Bulgaria 5.3 1.7 8.6
Pakistan 225 65 365 Egypt 5.1 1.7 8.2
Indonesia 219 57 368 Serbia and Montenegro 4.9 1.6 8.0
Egypt 194 64 310 Belarus 4.8 1.5 7.9
Ukraine 193 61 3`18 Lithuania 4.7 1.5 7.7
Japan 148 50 236 Latvia 4.3 1.4 7.2
Germany 133 42 218 Afghanistan 4.3 1.5 6.9

RESPIRATORY + CARDIOVASCULAR

PER
COUNTRY TOTAL LOW HIGH COUNTRY LOW HIGH
MILLION
India 3 167 472 5 567 Ukraine 7.6 1.7 13.2
China 2 984 729 5 003 Lesotho 7.5 2.0 12.6
United States 761 189 1 284 Egypt 7.3 1.9 12.1
Russian 488 117 839 Georgia 7.2 1.9 12.1
Japan 399 108 659 Korea Dem. People’s Rep. 6.9 1.9 11.4
Pakistan 384 74 659 Afghanistan 6.8 1.9 11.1
Indonesia 373 52 671 Bulgaria 6.7 1.6 11.5
Egypt 276 72 457 Serbia and Montenegro 6.4 1.5 11.0
Brazil 255 54 447 Uzbekistan 6.3 1.7 10.4
Germany 254 60 436 Eritrea 6.1 1.2 10.8
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
In prior modelling (Shindell et al. 2012; UNEP/
WMO 2011), each million tonne of methane was
estimated to lead to 340 premature deaths per
year due to ozone (220–490; central 66 per cent
confidence interval). The new value of ~1 400 is
roughly four times greater. In large part, the increase
in the estimated impact stems from the use of the
updated epidemiology based on the extended
ACS CPS-II cohort relative to the epidemiology
based on the earlier version of this same cohort
(Jerrett et al. 2009). The newer analysis of this
cohort includes not only a greater impact of ozone
on respiratory mortality but also the addition of
the impact of ozone on cardiovascular mortality.
Additional studies on the links between ozone
exposure and cardiovascular mortality would
be useful as the evidence base for this outcome
is less robust than that for respiratory mortality.
For example, the US Environmental Protection
Agency recently published an Integrated Science
Assessment for Ozone in which it noted that
a limited number of epidemiologic and animal
toxicological studies contribute new evidence
characterizing the relationship between long-
term ozone exposure and cardiovascular health
effects. That evidence includes epidemiologic
evidence that long-term ozone exposure may
be associated with blood pressure changes or
hypertension among different life stages or those
with pre-existing disease, as well as toxicological
evidence of inflammation, oxidative stress,
and impaired cardiac contractility in rodents
following long-term ozone exposure. Their
assessment concluded that the evidence of the
relationship between long-term ozone exposure
and cardiovascular health effects is suggestive
of a causal relationship, but not more definitive.
Given the differences in their evidence bases, this
assessment presents the results for respiratory
and cardiovascular impacts separately.
Additionally, the newer epidemiology uses the
metric of the maximum daily 8-hour average
over the entire year whereas the older one was
based on the warm-season maximum daily
8-hour average. As methane is particularly
important during winter months, the response
to methane is virtually certain to have been
enhanced by the change in the metric. We note
that the relationship between ozone exposure
and premature death, including exposure timing
and ozone-attributable risk, is still an active
area of scientific investigation. Total deaths per
million tonnes of methane attributable solely to
respiratory impacts are, as noted, 740 (460–990;
95 per cent confidence interval). Hence the results
here for respiratory impacts are roughly double
those found in the earlier work. Previous analyses
57
found a 32–50 per cent increase in respiratory
only impacts using the larger, extended cohort
relative to the older risk estimates (Seltzer et al.
2018) based on observed ozone rather than the
response to methane. This suggests that much
of the increase may be related to the metric
change, with an additional portion potentially
due to the use of newer models of atmospheric
chemistry and new emissions datasets for non-
methane species that affect ozone.
It should be noted that the larger impact of
ozone on health has been reported in several
previous studies. Malley et al. (2017) used the
new health exposure relationships (Turner et al.
2016) along with modelled ozone distributions,
and found a 125 per cent increase in respiratory
deaths attributable to ozone exposure in 2010
compared to previous estimates – 1.04–1.23
million deaths compared to 0.40–0.55 million.
Using observed ozone data for the China,
Europe and the United States, where coverage
is fairly complete combined with statistical
infilling of missing data, another recent study
showed that by including both respiratory and
cardiovascular impacts of ozone based on the
Turner et al. (2016) epidemiology impacts were
153 per cent larger across the three regions
than the values with the prior epidemiology,
which included respiratory effects only (Seltzer
et al. 2018). Hence the larger impacts are robust
despite the tendency of models to overestimate
surface ozone concentrations. Further to this,
a bias-adjusted model recently reported total
worldwide ozone-related premature deaths of
1.0 ± 0.3 million (Shindell et al. 2018). The value
for respiratory-related premature deaths due to
ozone was 0.6 ± 0.2 million for 2010, and 1.0
± 0.3 million without bias adjustment, the latter
being consistent with the value reported by
Malley et al. (2017).
3.4.2 HEAT-RELATED MORTALITY
Exposure to heat, even at temperatures that are
moderately hot, can compromise the body’s ability
to regulate its internal temperature, potentially
resulting in heat exhaustion, hyperthermia,
worsening of chronic conditions and heatstroke,
leading to temperature-related deaths
(Sarofim et al. 2016; Gasparrini et al. 2015;
Lee et al. 2014). Small temperature changes
at mild or moderate temperatures can have
larger health impacts than changes at extreme
levels, such as during heatwaves or cold snaps,
since they occur more frequently (Wellenius et
al. 2017; Sarofim et al. 2016; Gasparrini et al.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
2015). Heat exposure is currently the primary
weather-related illness in the United States,
with millions affected annually (Luber and
McGeehin 2008). Cold-related mortality is not
addressed here given its complexities and
uncertainties. In particular, it is unclear to what
extent empirically derived cold-related impacts
are likely to reflect causal mechanisms given
that other confounding affects such as influenza
occur with a similar seasonality. Exposure to
both heat and high humidity contribute to health
impacts (Buzan and Huber 2020; Li et al. 2020).
Though humidity changes are not explicitly
accounted for here, since the epidemiological
exposure–response relationships are based
on temperature alone, they are implicitly
included to some extent as humidity changes
are highly correlated with global-scale surface
air temperature changes that are included
in the long-term epidemiological studies.
Epidemiological studies have also reported that
despite the known impacts of high humidity on
health, there is little evidence that using heat
indices that account for both temperature and
humidity provides a better fit to the observed
mortality data (Armstrong et al. 2019).
In the case of premature deaths owing to
heat exposure, impacts based on the climate
response simulations (C1 and C2) are analyzed
here. The analyses are focused on China and the
United States as detailed generalized exposure–
response relationships have been derived for
these two nations from a substantial body of local
epidemiological evidence. Most prior studies
have relied on city-specific epidemiological
studies that created local exposure response
functions for the influence of heat exposure on
human health (Weinberger et al. 2017; Schwartz
et al. 2015; Honda et al. 2014). Based on the
available city-specific exposure–response
functions for the United States, generalized
functions that were shown to be applicable
nationwide were recently developed (Shindell et
al. 2020). These capture the geographic variation
in the observed epidemiological functions
by including a dependence on local climatic
conditions, which may also incorporate at least
a portion of localized differences in humidity.
As climate responses are evaluated for one to
three decades after emissions reductions take
place, the generalized exposure–response
functions to the 2030 population were applied
together with baseline mortality data using the
sources described above for ozone-related
deaths. In this instance, based on the underlying
epidemiological studies, impacts of heat were
applied as an altered all-cause risk rather than
58
cause-specific as for ozone. Hence all-cause
all-age mortality projections were used along
with projected population changes, again
with the geographic distribution of population
unchanged. Based on results described in Ma
et al. (2014), generalized equations have been
created for China analogous to those developed
previously for the United States (Shindell et
al. 2020). Briefly, using data from 15 cities in
China, relative-risk curves were created from the
epidemiological exposure–response functions
using a second-order polynomial following the
equation RR = 1+ aT2 + bT. The relationship
between the coefficients a and b and the local
city mean summer temperature (MST) was
then derived by linear regression to create a
generalized exposure–response function that
captured the sensitivity of this function to local
climatic conditions, as done for the United
States. The result is a generalized polynomial
function with the equation:
RR = 1 + as × (MST – ai )T2 + bs × (MST – bi )T
where as and bs were the slopes of the linearly
regressed coefficients versus mean summer
temperature and ai  and bi were the intercepts
of the same regressions. The mean summer
temperature is unique for each city, and T is the
temperature in degrees centigrade above the
optimum temperature defined as the value at
which relative risk is equal to 1 – set to the 84th
percentile of temperatures at a given location.
For the United States, the coefficients are:
as = -0.0014, ai = 30.9, bs = 0.005, bi = 26.7.
For China, the coefficients are:
as = 0.0037, ai = 21.0, bs = -0.0045, bi = 23.8.
Averaged over all available cities, the generalized
equations yield results for current conditions
(~2010) that are relatively close to those
obtained using the local epidemiological results,
with values 9 per cent higher for the 15 Chinese
cities and 7 per cent lower for the 10 cities in
the United States. Hence these generalized
equations appear to provide valuable insights
into national level impacts of heat exposure.
In the analysis conducted for this assessment,
the above generalized function at the grid cell
level across the China and the United States
were applied. As was the case for ozone-related
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
deaths, epidemiological studies do not provide
distinct results according to gender. In the
primary calculations, adaptation is not included.
In a sensitivity study, the value for the mean
summer temperature was allowed to change
between the C1 and C2 simulations in order
to represent human adaptation to long-term
seasonal temperature changes. In a warming
climate, for example, United States’ cities in
cooler regions take on more of the character of
warmer cities, becoming less sensitive to heat
extremes. In contrast, the optimum temperature
is kept fixed at the value from simulation
C1 to represent the influence of short-term
temperature excursions (see Shindell et al., 2020
for additional discussion of adaptation).
It was found that the 50 per cent reduction in
methane leads to 2 800 (1 700–3 800; 95 per
cent confidence interval) fewer annual heat-
related deaths in the United States and 14 000
(6 000–21 000; 95 per cent confidence interval)
fewer in China based on the climate response
averaged over one to three decades following
emissions reductions and accounting for
the uncertainty in that national level climate
response. For comparison, the total annual
ozone-related deaths in these countries are
10 200 for the United States and 40 100 for
China. Hence ozone-related reductions in
deaths are roughly factors of 3–4 larger. As the
ozone-related reductions begin immediately
rather than a decade later, they have a larger
overall impact on mortality rates in these two
countries. Nonetheless, this analysis, especially
the results for China, suggests that the human
health benefits attributable to reduced climate
change are also substantial and that they could
contribute a large fraction of total health benefits,
in particular in countries with low ozone pollution
levels. Based just on these two countries for
which data are available, each million tonne of
methane emissions mitigation leads to about 125
fewer premature heat-related deaths. The ozone-
related deaths in these two countries make up
about one quarter of the global total, suggesting
that the global total might be roughly 500 fewer
premature heat-related deaths per million tonnes
of methane if the ratio of China+United States to
global heat-related deaths is similar to that for
ozone-related deaths.
Potential impacts of adaptation by accounting
for the change in summer mean temperatures,
as described above, were examined. This led
to a reduction in the United States annual heat-
related deaths of ~100, or about 2 per cent
(0–5 per cent across models). This suggests
only weak effects for at least this type of
59
adjustment to mean climatological changes.
Warmer cities in the United States experience
less increase in the risk of heat-related mortality
per degree warming than cooler cities, so that
acclimatization to warmer temperatures leads
to reduced risk. In China, however, the opposite
is true in that warmer cities experience greater
risk per degree of warming. Hence this method
of accounting for adaptation was not applied to
China. Presumably, the difference between these
countries is related to the greater wealth in the
United States that allows nearly complete use of
air conditioning in warm climates, making them
better adapted to heat extremes than cooler
climates, whereas air conditioning ownership in
China has only expanded rapidly very recently,
after the epidemiological studies on which the
generalized equation for China is based were
performed. Given the minimal changes in the
United States and lack of an adequate method
for China, hereafter only the values without
accounting for possible adaptation are reported,
though it is noted that a broader exploration of
potential adaptation found that it might reduce
impacts roughly by half (Shindell et al. 2020;
Anderson et al. 2018; US EPA 2017).
An additional estimate of wider scale heat-related
mortality impacts is based on the national-level
responses to national average warm-season
temperature changes reported in a recent
study (Lee et al. 2019). This study reported an
increase in national level risk for heat-related
mortality as a function of national level warming
based on averaging across available local
datasets, typically from several urban areas
within each country. Regional averages were
also provided for most parts of the world, with
the significant exceptions of Africa and South
Asia. The responses were examined using this
much rougher relationship with the results from
our spatially explicit response to the distribution
of daily temperature values in China and the
United States. The national-average seasonal-
mean method produced an impact that was 55–
57 per cent of that found in the more detailed
analyses of China and the United States. Based
on the consistency across those countries, and
assuming the more spatially and temporally
explicit method was the more accurate, all
national results from the other method were
normalized by dividing by 0.56.
Using this normalized national-average
seasonal-mean method, the total across
regions with data is 53 000 (39 000–74 000
across the three models). This corresponds
to about 390 avoided heat-related deaths per
million tonnes of methane emissions reduction.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
Given the exclusion of Africa, except for Egypt,
which is included as part of the Middle East in
this analysis, South Asia and this value appears
to be fairly consistent with the estimate of 500
deaths per million tonne of methane from the
extrapolation of China and the United States
results based on the ratio of the ozone-related
impacts in those two countries to the global total.
More specifically, the World Health Organization
estimated worldwide mortality effects of heat
exposure, based on data from Japan alone, and
reported that 37 per cent of premature deaths
due to heat were projected for 2030 in Africa and
South Asia (WHO 2014). Applying this same ratio
to the results in this analysis would imply ~600
deaths per million tonne of methane.
Though both large-scale estimates are relatively
crude and thus confidence in the precise values
is low, they provide an indication of a plausible
level of worldwide heat-related mortality impacts,
adding to a growing body of evidence for such
widespread effects (Carleton et al. 2020; Lee et
al. 2019; WHO 2014). It should also be noted that
the Lee et al. (2019) study reported that impacts
were fairly linear with warming, so that the scaling
approach used throughout this assessment
appears reasonable for this impact as well,
although due to data limitations that specifically
could not be examined in this analysis. These
results suggest that there are, very roughly, three
times more ozone-related deaths than heat-
related ones. Additionally, ozone-related deaths
would decrease immediately in response to
reductions in methane emissions, with ozone-
related deaths evolving rapidly over time following
the ~12 year adjustment time for methane,
whereas heat-related deaths would change
more slowly as climate responds, with the values
estimated here occurring one to three decades
after emissions changes.
3.4.3 MORBIDITY AND LABOUR PRODUCTIVITY
The effect of methane emissions reductions is
examined here on two non-fatal health impacts
in many countries:
1. hospital admissions due to respiratory illness
resulting from ozone exposure; and
2. visits accident and emergency departments
due to asthma resulting from ozone exposure.
The change in hours of lost labour due to extreme
heat is also quantified, but only in the United
States due to data limitations.
60
3.4.3.1 MORBIDITY AND LABOUR
PRODUCTIVITY METHODS
Changes in respiratory hospital admission are
evaluated for the population aged 65+ based
on decreases in ozone exposure in response to
methane emissions reductions. Estimates are
based on the epidemiological studies of three
groups of investigators, in Canada, Europe and
the United States, covering 149 cities globally
under the Air Pollution and Health: a European
and North American Approach (APHENA) project
(Katsouyanni et al. 2009). Hospital admissions
data were collected from medical records for
all respiratory diseases, conditions or infections
(International Classification of Diseases,
Ninth Revision (ICD-9) codes 460–519). This
global effort showed a robust relationship
between ozone exposures and respiratory
hospital admissions for this age group, without
distinguishable geographical difference. This
relationship was therefore applied worldwide. As
with mortality, the epidemiology does not clearly
indicate whether there is any differentiation in
risk by gender.
A well-established approach (Fann et al. 2018;
Heroux et al. 2015; WHO 2013) was used to
estimate respiratory hospital admissions.
AF = 1 – exp -ß∆[0 ]
3
∆Hospital admissions = y0 × AF × Population
where β is the exposure-response function,
that is the slope of the log-linear relationship
between unit change in exposure and respiratory
hospital admission, developed from APHENA,
AF is the attributable fraction of hospital
admissions associated with ozone exposure, y0
is the baseline hospitalization rate of respiratory
diseases among the 65+ age group, and
population is the count of 65+ year-olds at each
0.5º x 0.5º grid box. This approach is similar to
the method used to calculate mortality, except
that no lower exposure threshold is applied for
morbidity. Given the short-term nature of the
link between respiratory hospital admission
and ozone exposure, responses in hospital
admissions on a daily basis were estimated and
the total summed over a year.
National-level variations result from the uneven
distribution of ozone changes, country-specific
hospitalization admission or discharge rates for
those aged 65+ hospitalized due to respiratory
diseases, and population distributions of 65+
year-olds. The exposure–response functions
studied in APHENA have used both penalized
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
spline and natural spline models. For this
analysis the average of the two exposure–
response functions was calculated, and a
relative risk of 1.006 (0.0008–1.012) used for
per 10 ppbv of ozone maximum daily 8-hour
average (i.e. β is 0.00063 (0.000087–0.0012)).
Data on the post-65 population were again
from GPW version 4 (CIESIN, 2016), the same
dataset used for mortality estimation. Baseline
hospital admission data are compiled from
multiple datasets. Efforts were made to include
all the available datasets representing a baseline
level of respiratory hospitalization. The baseline
rate of respiratory hospital admissions for the
United States was obtained from Environmental
Benefits Mapping and Analysis Program –
Community Edition (BenMAP-CE) (US EPA
2015), which calculated the baseline respiratory
hospital admission rate by age-groups from the
Healthcare Cost and Utilization Project (HCUP)
2014. This is also the only available dataset
that reports age-specific rates. The all-age rate
for respiratory hospital admissions for China
was obtained from China’s 2016 Health and
Family Planning Statistics Yearbook (NHFPC,
2016) while the all-age hospital discharge rate
for respiratory diseases for European countries
came from Eurostat (Eurostat, 2016). Hospital
discharge rates here, which are virtually the
same as the hospital admission rates, include
“finalisation of treatment, signing out against
medical advice, transfer to another healthcare
institution, or because of death”. To transform
the all-age rate to the 65+ rate in China and
Europe, a ratio of 3.30:1 was applied, which
was the average ratio of hospital admissions
between those aged 65+ and all-age groups that
was calculated from the United States dataset.
Finally, the annual hospital admission rates
were divided by 365 to obtain a daily average
hospitalization rate. Given the limited availability
of data, this compiled dataset of 35 countries
from multiple sources was not ideal and may
have introduced heterogeneity. In particular, it
assumes the ratio between those aged 65+ and
all-age groups in China and Europe is the same
as that in the United States. It also assumes that
there are no seasonal variations in respiratory
hospital admissions over the year. The impacts
of these heterogeneity and assumptions are,
however, likely to be significantly lower than
the uncertainty associated with the exposure–
response function. Ozone responses are based
upon the modelling in simulations 1 and 2.
Asthma-related accident and emergency
department visit changes were evaluated for
the entire population based upon decreases
61
in ozone exposure in response to methane
emissions reductions. Results are based upon
the average of epidemiological relationships
across three meta-analyses of existing literature
that reviewed 22, 26 and 50 studies (Orellano et
al. 2017; Zhang et al. 2016; Zheng et al. 2015).
These reviews cover studies in the Asia, Europe
and the United States. The approach developed
in Anenberg et al. (2018) was used to estimate
the global asthma accident and emergency
department visits due to ozone exposure.
AF = 1 – exp -ß∆[0 ]
3
∆Hospital admissions = y0 × A&EV × AF × Population
where, again, β is the exposure–response
function, that is the slope of the log-linear
relationship between unit change in ozone
exposure and asthma emergency room visits,
AF is the attributable fraction, y0 is the country-
specific baseline prevalence rate of asthma for
the all-age population, accident and emergency
department visits (A&EV) is the fraction of
individuals with asthma visiting accident and
emergency departments in each country, and
population is the all-age population count in
each grid box. Note that although accident and
emergency department visits are a short-term
morbidity impact, estimation is usually based on
the relationship with an annual change of ozone
exposure. As such, an annual average of ozone
exposure changes was calculated using annual
maximum daily 8-hour average and the above
method applied.
Exposure response functions were applied
from the three meta-analyses respectively:
Orellano et al. (2017): 1.03 (1.01–1.06); Zhang
et al. (2016): 1.05 (1.04–1.07) and Zheng et al.
(2015): 1.02 (1.01–1.02). An average of the three
results for estimated asthma-related accident
and emergency department visits was then
calculated. Annual country-specific baseline
asthma prevalence rate for the all-age group
was collected from the GBD 2018 (Global
Burden of Disease Collaborative Network, 2018).
The A&EV for available countries compiled by
Anenberg et al. (2018) was used, which includes
multiple national data sources. The final results
here are for 56 countries and territories, due
to data availability in A&EV . The gridded all-
age population was again from GPW version
4 (CIESIN 2016). Ozone responses were again
based on the modelling in simulations 1 and
2. As both ozone morbidity impacts are based
upon the maximum daily 8-hour average, and
prior studies have shown that a sum of daily
maximum daily 8-hour averages is very close
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
to the annual maximum daily 8-hour average
(Anenberg et al. 2018), the analysis presented
previously regarding the linearity of the annual
maximum daily 8-hour average (Figure 4.3)
applies to these endpoints as well and hence
hereafter it is assumed these impacts can be
represented by a linear scaling from the results
explicitly modelled in this assessment.
Labour losses due to exposure to extreme heat
are evaluated using two distinct methods. The
first is based on an empirically established
exposure–response function for the United
States (Graff Zivin and Neidell 2014). To evaluate
this exposure–response relationship, nationally
representative survey data from 2003–2006 and
daily weather observations from roughly 8 000
weather stations were used to investigate how
Americans allocate their work and leisure time as
a function of ambient temperatures. The authors
found a statistically significant approximately
linear decrease in the time allocated to labour
with increasing temperatures above a threshold
of about 29° C for high-risk sectors. This
relationship has been used in several recent
studies of the United States (Hsiang et al. 2017;
USGCRP 2018). In the calculations for this
assessment, the value of time lost is calculated
using county-level annual employment, covering
15–64 year-olds, and annual average weekly
wages from the Quarterly Census of Employment
and Wages, U.S. Bureau of Labor Statistics. The
North American Industry Classification System
was used to determine the number of workers in
high-risk industries/sectors – largely those that
cannot readily be air-conditioned: agriculture,
forestry, fishing and hunting; construction;
manufacturing; mining; transportation: and
utilities. The 2016 fraction of workers in high-
risk industries was used for each county. The
analysis covers the contiguous United States,
that is, excluding Alaska and Hawaii. Surface
temperature changes in response to methane
emissions reductions are based upon the daily
temperature values modelled in simulations
C1 and C2. Ninety years of data were used,
years 11–40 from each of the three models that
performed the C1 and C2 simulations, giving
adequate data to establish a clear signal despite
the noise inherent in daily temperatures.
The second method is based on a publication
(Knittel et al. 2020) that appeared after the
modelling to support this assessment was already
largely completed. The relationship between
heat and labour productivity in this study has
9. Kilopascal = 1000 pascals
62
the advantage of being applicable globally, but
it relates productivity to relative humidity as well
as temperature and daily humidity data was only
available from an extension of the simulations
performed for 20 years with one model as the
need for this output had not been anticipated
(thought the study that brought this method to
our attention relied upon older methods). In the
analysis using this method, following Knittel et
al. and Kjellstrom et al. (2018), daily mean and
max surface air temperature and daily mean
relative humidity were used to represent the wet
bulb globe temperature (WGBT) or heat index
using the following equation:
WBGT = 0.567×T + 0.393×Vapour + 3.94
with T as either the daily mean, daily max or the
average of those (daily halfmeanmax) 2-metre
air temperature, in ºC, and Vapour as the
water vapour partial pressure in kilopascals9.
This equation is representative for light wind
conditions and moderately high heat radiation
levels typically experienced by outdoor workers.
The WBGT minus 4 is used to represent
conditions experienced by indoor workers
(Knittel et al. 2020).
Vapour was calculated using the daily means of
relative humidity (RH) and T using:
17.27×Tavg
Vapour = RH ×6.105 ×exp (
— ——— )
100 237.7×Tavg
The WBGT was then related to work ability (WA)
following Knittel et al. (2020) based on the work
of Bröde et al. (2017) and Kjellstrom et al. (2014):
WA =0.1+0.9/(1+(WBGT/α1)α2)
where and are unitless coefficients derived for
the three different work intensity categories: light,
24.64 and 22.72; medium, 32.98 and 17.81; and
heavy work, 30.94 and 16.64. Light work includes
all the service sectors, such as retail sales,
wholesale trade, hotels and restaurants, post
and telecoms, financial services and insurance,
and public administration, and is assumed to be
indoors. Medium work includes such sectors
as food, textile and wood industries, machinery
and electronic equipment, and transport and is
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
likewise assumed to be indoors (i.e. in a shaded
or air conditioned space rather than direct
sunlight). Heavy work includes the agriculture,
forestry and fishery, extraction and construction
sectors and is assumed to be outdoors.
Work ability is assumed to be limited to 0.1 to 0.9
(in portion of an hour) to allow for some minimal
work and acknowledge that breaks are needed
even under optimal conditions. The calculated
fractional work loss, WL=1-WA, is then multiplied
by the working population, adults aged 15–64, in
each country in each work category to get the
annual total labour loss due to heat. The lost
hours are based on 12 months multiplied by 30
days per month multiplied by 12 hours per day,
with a distribution of work that has 4 hours each
at daily maximum, daily half mean max and daily
mean temperatures; these assumptions have
been shown to provide a good approximation
of the hourly temperature distribution during the
day (Kjellstrom et al. 2018). This leads to 4 320
potential work hours per year. The country-
specific working population in 2018 for each
labour category was based on data from the
International Labour Organization (ILO), which
includes gender-specific information (ILO,
2020). As in Knittel et al. (2020) and Kjellstrom
et al. (2018), the relationship between WA and
WBGT is assumed to be applicable worldwide
although the underlying data is largely from
more developed nations.
3.4.3.2 MORBIDITY AND LABOUR
PRODUCTIVITY RESULTS
The ozone decreases in response to reductions
in methane emissions lead to an estimated
reduction of 855 annual hospitalizations for
persons aged 65+ per 10 Mt of methane in
countries for which sufficient data is available.
Among those nations, the largest values are
seen in China, followed by the United States
(Table 3.4; Figure 3.10). Effects are larger than
in either of those nations, however, for Europe.
Note that data is unavailable to allow these
effects to be quantified outside of China, Europe
and the United States. In addition, the overall
uncertainty range for this impact is quite large,
±84 per cent, owing primarily to the uncertainty
in the exposure–response relationship.
The ozone decreases in response to reductions
in methane emissions also lead to a decrease
in asthma-related accident and emergency
department visits, with a response of 43050
63
fewer cases per 10 Mt methane emission
cuts. Note that in comparison with respiratory
hospitalizations, there is much more data
available and so the impact on asthma-related
accident and emergency department visits is
quantified for many more nations (Table 3.5a;
Figure 3.11). In addition, the epidemiological
relationship has a much narrower uncertainty
range, so that the overall uncertainty in the
change in number of accident and emergency
department visits is much smaller at ±37 per cent.
Among the many countries for which impacts
were estimated, benefits in India are far larger
than in any other nation, with more than half the
avoided cases taking place there. China ranks
second and the benefits distinctly larger than in
countries ranking third and lower. Several other
countries do see substantial benefits, however,
including both developed ones such as the
Republic of Korea, Japan and the United States
as well as developing countries including Brazil,
Ethiopia and Nigeria, all of which see more than
500 fewer accident and emergency department
visits per 10 Mt of methane emissions decrease.
As described above, the change in labour losses
due to exposure to extreme heat was calculated
for the United States only using the first method
applied to multiple models. Based on the
simulations of the climate response to methane
emissions changes, the cooler climate resulting
from the methane emission reduction between
simulations C2 and C1 leads to 13.6 million (2.2–
18.2; 95 per cent confidence interval) fewer lost
work hours annually in the United States. This
translates to approximately 1.0 (0.2–1.4) million
hours per 10 Mt of methane.
The second method provided work loss
estimates for most of the world, but in this case
from a single model only. It is noted that the
second method finds a much larger loss for the
United States, namely 44 million fewer lost work
hours annually in the United States. The second
method is based on the temperature changes
simulated in the GISS model and using the first
method that model produced an estimate of just
4.8 million lost work hours rather than the 13.6
million lost work hours for the multi-model mean.
Hence a multi-model mean with the second
method would be likely to find larger impacts
than the results based on the GISS model alone.
Using the GISS climate simulations, the second
method finds that the cooler climate resulting
from the methane emission reduction between
simulations C2 and C1 leads to 55 billion lost
work hours annually (±38 billion based on the
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 64

GISS model’s interannual variability over the
20 year simulation). Using this global method,
effects tend to be greatest in low latitude, humid
countries in which the daily heat index or WBGT
frequently reaches high values (Table 3.5b;
Figure 3.12). This corresponds to ~410 million
lost work hours per million tonne of methane
emitted, assuming responses are again quasi-
linear. Employment within those sectors of the
economy that are affected by heat exposure
varies across gender, leading to disparities in the
impacts for men and women in most countries.
While in a few, such as Angola, Uganda,
Cambodia, Viet Nam, Azerbaijan, Brazil and
Kenya, there is nearly an even split between lost
hours for men and women, in most the majority
of the lost hours are for male labour, with female
labour losses often accounting for only 20–35
per cent of the total (Table 3.5b). Nepal stands
out as the only country with substantial losses,
millions of hours lost in response to a 134 Mt
decrease in emissions, for which the loss of
labour is substantially higher amongst women,
59 per cent, while globally 66 per cent of impacts
are felt by men and 34 per cent by women.
In a related analysis, the ILO reports that
increased heat projected for 2030 could bring
productivity losses equivalent to 80 million jobs
(ILO 2019). Converting 80 million jobs to hours
lost, assuming 50 weeks of 40 hours are worked
per year, this translates to roughly 160 billion
lost work hours annually. Their analysis hence
similarly identifies the potential for large impacts.
It also indicates substantial non-linearities in
response, with impacts in 2030 nearly 10 times
greater than those in 1995 despite 2030 warming
being only roughly double the 1995 value. This
makes it difficult to compare the results here
directly with the ILO’s, and also suggests that
assuming a linear response as in this assessment
provides only a very rough measure of impacts
and that the non-linearity of this response merits
further characterization.

Table 3.4 Respiratory hospital admissions due to ozone exposure per 10 Mt methane emissions for countries with
available data

Note: other countries are likely have impacts as well, though these are not quantified here.

COUNTRY CASES COUNTRY CASES

China 247 Norway 5.7

United States 145 Denmark 4.7
Italy 96 Czechia 4.5
Germany 78 Serbia and Montenegro 4.5
France 59 Slovakia 3.0
Spain 51 Finland 2.5
United Kingdom 45 Croatia 2.4
Poland 21 Slovenia 2.4
Greece 13 Ireland 2.1
Romania 9.9 Lithuania 1.2
Belgium 9.9 Estonia 0.9
Portugal 8.3 Cyprus 0.7
Sweden 8.1 Malta 0.5
Bulgaria 7.2 Luxembourg 0.3
Austria 7.2 Latvia 0.3
Netherlands 6.7 Iceland 0.2
Hungary 6.2 TOTAL 855

The uncertainty range is ±84 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 65

Figure 3.10 Change in respiratory hospital admissions for people aged 65+ due to ozone exposure change resulting
from a 134 Mt change in methane emissions, China, Europe and United States

Note: Based on the multi-model mean ozone change. Values are for China, the US, Europe and China
only; although other countries are likely have impacts, these have not been quantified here due to
data limitations – grey indicates no quantification.

Source: UNEP and CCAC

Figure 3.11 Change in asthma-related visits to accident and emergency departments due to ozone exposure change
resulting from a 134 Mt increase in methane emissions, all ages, selected countries

Note: Based on the multi-model mean ozone change. Values given for countries with available
underlying data – grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 66

Figure 3.12 Change in lost labour hours due to heat exposure resulting from a 134 Mt increase in methane emissions,
selected countries, million hours

Note: Values given for countries with available underlying data and statistically significant temperature
responses – grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 67

Table 3.5a. Asthma-related accident and emergency department visits due to ozone exposure, all age groups, per
10 Mt of methane emissions

COUNTRY CASES COUNTRY CASES

India 23 000 Australia 110

China 7 200 Norway 108

United States 1 500 Jordan 95

Brazil 1 100 Italy 83

Ethiopia 1 100 United Arab Emirates 81

Nigeria 1 100 Poland 69

Viet Nam 860 Sweden 64

Turkey 580 Lebanon 63

Russia 570 France 63

Korea, Rep. 530 Costa Rica 56

Japan 510 Czechia 51

Argentina 410 Romania 47

Algeria 380 Oman 46

Spain 380 Belgium 46

Colombia 360 Tunisia 40

Thailand 345 Kuwait 38

Peru 308 Hungary 35

Morocco 261 Austria 26

Venezuela 246 Finland 23

Canada 224 Croatia 23

Ukraine 220 Uruguay 18

Philippines 161 Bulgaria 17

Germany 150 Singapore 17

Ecuador 133 Slovakia 15

United Kingdom 121 Latvia 12

Netherlands 114 Slovenia 12

Malaysia 113 Lithuania 10

Chile 111 TOTAL 43 050

The uncertainty range is ± 37 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 68

Table 3.5b. Reduced labour losses due to heat exposure resulting from a 134 Mt decrease in methane emissions for
the nations with the 15 largest number of hours lost

COUNTRY LOST WORK HOURS (MILLION) LOSSES FOR WOMEN (%)

India 12 000 22

China 6 700 39

Indonesia 5 900 32

Brazil 4 900 44

Viet Nam 2 300 45

Ethiopia 2 200 41

Philippines 1 700 19

Bangladesh 1 600 32

Thailand 1 400 42

Tanzania 1 200 47

Uganda 1 200 51

Pakistan 1 200 23

Colombia 1 100 36

Kenya 1 100 50

Mexico 800 17

GLOBAL TOTAL 55 000 34

The uncertainty range is ± 70 per cent.

3.5 CROP IMPACTS: METHODS AND
RESULTS
Methane also plays a significant role in reducing
crop yields and the quality of vegetation. Ozone
exposure is estimated to result in yield losses
in wheat, 7.1 per cent; soybean,12.4 per cent;
maize, 6.1 per cent; and rice, 4.4 per cent for
near present-day global totals (Mills et al. 2018;
Shindell et al. 2016; Avnery et al. 2011a). Impacts
on these four crops for which robust ozone
concentration-response functions are available,
were evaluated using an empirical crop model
based on statistical relationships for the impacts
of temperature, precipitation, carbon dioxide
concentrations, and ozone, rather than plant level
simulations (Shindell et al. 2019). Wheat, maize
and rice responses to changes in meteorological
variables are based on a meta‐analysis of more
than 1 000 modelling studies (Challinor et al.
2014), incorporating relationships observed
in field studies. Responses for soybeans to
temperature are based upon a separate study
(Zhao et al. 2017) as those were not included
in the meta-analysis. Separate temperature
response coefficients for wheat, maize, and
rice are included according to temperate or
tropical conditions. For ozone, the M7 and
M12 exposure metrics, the mean 7‐ or 12‐hour
daylight exposure during the growing season,
depending on the crop, are used. These affect
yields based on the response reported in field
studies (Wang and Mauzerall 2004), and have
been used previously for global crop modelling
(Van Dingenen et al. 2009). The growing season
is locally defined based on Van Dingenen et al.
(2009). Changes in temperature, precipitation
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
and ozone in response to methane emissions
changes are taken from the simulations
performed for this assessment. In addition,
a small amount of carbon dioxide is formed
when methane is oxidized. This carbon dioxide
response fertilizes crop growth, especially for
C3 plants such as wheat, soybeans and rice
with only very small effects for maize, a C4 plant
(C3 and C4 refer to the type of photosynthesis
used by the plant to create energy, with the
first carbon compound produced containing
either 3 or 4 carbon atoms). These effects are
included here based on relationships described
previously (Tebaldi and Lobell 2018). Crop
distributions for 2010 were taken from the Food
and Agricultural Organization of the United
Nations (FAO) data sets10 and are maintained at
those levels in all calculations of impacts – no
projected changes in crop areas are included.
It is noted that ozone has additional detrimental
impacts on ecosystems, including the loss of
protein content in grains, loss of forage quality
in grasslands, and losses in forest productivity
and carbon uptake.
The effects of ozone and temperature dominate
the response, with only minimal impact from
projected changes in precipitation or carbon
dioxide at large scales, as in prior analyses
(Shindell et al. 2019). The relative importance
of temperature change versus ozone change
varies across crops and locations. They are, for
example, similar in importance for maize, but
temperature has a much larger impact than ozone
on tropical wheat, whereas ozone dominates for
soybeans in most locations and is larger for rice
in many nations. Based on the quasi-linearity of
the responses of both ozone and temperature
to methane emissions changes, the modelled
impacts were converted to crop impacts per unit
methane emission.
The global totals for a 50 reduction in
anthropogenic-related methane increases, a 134
Mt decrease in methane emissions, are avoided
yield losses of 7.46 Mt of wheat, 2.23 Mt of
soybeans, 5.58 Mt of maize and 4.20 Mt of rice,
with uncertainty ranges of roughly 36 per cent
based on the Monte-Carlo sampling performed
previously (Shindell et al. 2019). National level
totals are reported in Tables 3.6a–3.6d and
shown in Figure 3.13. Losses in terms of total
tonnage are naturally heavily weighted towards
10. faostat.fao.org
69
nations with large baseline yields as those vary
across countries much more than the relative
yield changes due to methane emissions.
Relative yield losses tend to be greatest in low
latitude countries, predominantly those in South
Asia, the Middle East and Africa.
The total effects reported here, avoided yield
losses of ~20 Mt, are slightly smaller than the
value of 27 Mt in response to a very similar
methane emissions reduction that was reported
in prior modelling using just two models
(Shindell et al. 2012; UNEP 2011). This is at
first somewhat surprising given that these new
estimates include both ozone and climate
whereas the earlier results were for ozone alone.
As the ozone exposure–response functions have
not changed, this suggests that the multi-model
analysis has produced a slightly smaller crop-
related ozone response (M7 or M12) than in the
prior study. The range given in the prior work,
however, was very large spanning 7–69 million
tonnes, as the two models differed markedly. The
prior studies used the GISS and ECHAM models,
with nearly a factor of two larger response in
the latter and the current analysis reporting a
multi-model mean close to the earlier low-end
model (GISS). Using several models this time,
the current results provide a substantially better
constrained value that is well within the large
range in the earlier analysis. Differences may
also stem in part from changes in background
conditions in predominantly rural areas where
crops are grown.
As with health-related impacts, the linearity of
crop impacts was examined across the various
prescribed changes in methane concentrations.
Responses are again relatively linear in nearly
all cases at the national level for countries with
substantial impacts (Figure 3.14). Significant
deviations are seen for maize in China and
rice in Bangladesh, with weaker responses
to an increase than a decrease in methane in
the former and vice-versa in the latter. Even in
these cases, however, a linear approximation
yields a fairly close result over the 0–134 Mt
decrease range. Hence as with health impacts,
hereafter it is assumed that crop responses can
be well approximated with a linear scaling of
the multi-model mean response from the half
anthropogenic methane simulations.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 70

Table 3.6a Total (Mt) and relative (per cent) losses in yield of maize per 134 Mt methane emissions for countries
with losses greater than 0.040 Mt and the top 14 countries for relative yield losses

COUNTRY YIELD LOSS (MT) COUNTRY YIELD LOSS (%)

United States 2.345 Bangladesh 3.1

China 1.554 Nepal 3.1
Brazil 0.214 Jordan 2.9
India 0.147 Tajikistan 2.9
Mexico 0.134 Iraq 2.8
Egypt 0.115 Pakistan 2.7
Italy 0.107 India 2.6
France 0.078 Iran 2.6
South Africa 0.060 Afghanistan 2.3
Argentina 0.059 Uzbekistan 2.2
Romania 0.050 Egypt 2.1
Hungary 0.047 Syria 2.0
Indonesia 0.043 Turkmenistan 2.0
Canada 0.040 Angola 1.8
GLOBAL 5.580 GLOBAL 1.2

The uncertainties are ~40 per cent.

Table 3.6b Total (Mt) and relative (%) losses in yield of wheat per 134 Mt methane emissions for countries losses
greater than 0.075 Mt and the top 16 countries for relative yield losses

COUNTRY YIELD LOSS (MT) COUNTRY YIELD LOSS (%)

India 2.700 United Arab Emirates 5.9

China 1.324 Saudi Arabia 5.8
Pakistan 0.556 Brazil 5.7
United States 0.449 Oman 5.2
France 0.203 Paraguay 5.1
Australia 0.200 Qatar 4.6
Turkey 0.186 Bahrain 4.6
Egypt 0.148 Bolivia 4.5
Iran 0.142 Madagascar 4.3
Germany 0.139 Yemen 4.2
Russia 0.115 India 4.2
Argentina 0.085 Sudan 4.1
Saudi Arabia 0.083 Eritrea 3.6
United Kingdom 0.076 Tanzania 3.5
Canada 0.076 Niger 3.5
Ukraine 0.075 Myanmar 3.3
GLOBAL 7.460 GLOBAL 1.7

The uncertainties are ~30 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 71

Table 3.6c Total (Mt) and relative (%) losses in yield of rice per 134 Mt methane emissions (for countries with losses
greater than 0.020 Mt and the top 16 countries for relative yield losses

COUNTRY YIELD LOSS (MT) COUNTRY YIELD LOSS (%)

India 1.432 Uzbekistan 1.8

China 1.252 Nepal 1.7
Bangladesh 0.360 Tajikistan 1.7
Indonesia 0.173 Afghanistan 1.6
Myanmar 0.119 Iraq 1.5
Thailand 0.102 Kyrgyzstan 1.4
United States 0.085 Bhutan 1.4
Viet Nam 0.075 Pakistan 1.3
Pakistan 0.071 Turkmenistan 1.3
Japan 0.071 Kazakhstan 1.2
Korea, Rep. 0.061 India 1.2
Nepal 0.053 Bangladesh 1.1
Egypt 0.053 United States 1.1
Brazil 0.038 Iran 1.1
Philippines 0.038 Australia 1.0
Angola 0.027 Rwanda 1.0
GLOBAL 4.202 GLOBAL 0.8

The uncertainties are ~40 per cent.

Table 3.6d Total (Mt) and relative (%) losses in yield of soy per 134 Mt methane emission for countries with losses
greater than 0.010 Mt and the top 11 countries for relative yield losses

COUNTRY YIELD LOSS (MT) COUNTRY YIELD LOSS (%)

United States 1.054 Iran 3.4

Brazil 0.412 Iraq 3.3
Argentina 0.231 Nepal 2.8
Congo, Dem. Rep. 0.182 India 2.8
India 0.077 Zambia 2.7
Australia 0.061 Tanzania 2.6
China 0.047 Kazakhstan 2.5
Paraguay 0.030 Korea, Rep. 2.4
Gabon 0.016 Zimbabwe 2.3
Canada 0.012 Angola 2.2
Italy 0.010 Mozambique 2.2
GLOBAL 2.232 GLOBAL 1.7

The uncertainties are ~40 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 72

Figure 3.13 Annual

change in soybean, rice,
wheat and maize yields
due to ozone exposure,
climate and carbon
dioxide changes resulting
from a 134 Mt change
in methane emissions,
thousand tonnes

Note: Based on
the multi-model
mean changes –
grey indicates no
quantification.

Source: UNEP and

CCAC
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 73

Figure 3.14 Change in wheat, soybean, maize and rice yields versus change in methane amounts, selected countries,
per cent

Note: Responses are shown for countries with the largest impacts for each crop based on GISS-E2
simulations. The charts include points from individual simulations along with linear fits for each
country. Negative losses represent gains.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
3.6 MONETIZATION OF IMPACTS:
METHODS AND RESULTS
Both market and non-market costs associated
with the impacts described above are evaluated.
All values are in constant 2018 US dollars. Non-
market values include those associated with
premature death and many of the effects of climate
change. Market costs include direct spending
on health care and the effects of environmental
changes on labour productivity.
Monetized benefits associated with avoided
mortality are evaluated using a willingness-to-pay
(WTP) measure of the value societies place upon
reduced risk of premature death. This measure
is often referred to as the value of a statistical
life (VSL) though it is in fact an expression of
the value that people affix to small changes in
mortality risks in monetary terms rather than
the value of any individual’s life. Health literature
often uses disability adjusted life years (DALYs),
which are arguably more informative since they
incorporate the age of the affected individuals
and years with disabilities as well as premature
deaths. Monetization is virtually always based
on VSL, however, which is a better-established
metric in the economics literature (Viscusi and
Aldy 2003). Here the societal willingness-to-
pay based is evaluated on national level gross
domestic product (GDP) using data from the
World Bank based upon national accounts data
(World Bank, 2019). Gross domestic product
values are used for 2018 except in the few cases
for which they are unavailable in which case the
most recent year is used – 2017 for Iran and 2011
for Eritrea. The WTP here is based on the mean of
26 peer-reviewed studies evaluated by the United
States Environmental Protection Agency (US EPA
1997) and used by that agency to derive its official
recommended VSL of US$ 7.4 million in 2006 US
dollars. That value is then inflated to represent the
year 2018 using an economic growth rate of 2.6
per cent per year and an elasticity for WTP with
income growth for any specific country of 0.4, as
in prior studies (Shindell et al. 2012; UNEP 2011).
To account for the fact that WTP is a function of
income across countries, monetized benefits are
evaluated using both the same elasticity of 0.4,
a value typically used to account for differential
income levels within a country or region, and
also using an elasticity of 1.0 as recommended
by Hammitt and Robinson (2011) for across
country analyses and used in the work of OECD
(2018) and the World Bank and the Institute for
Health Metrics and Evaluation (IHME) (2016) for
both high-to-low income and low-to-high income
comparisons. In other words, in this analysis, for
country a,
74
VSLa = VSLUS x (Incomea/IncomeUS)ß
where ß is either 1.0 or 0.4. Both sets of values
are presented (Tables 3.7a and 3.7b).
The results for ozone-related deaths show that
national level values are greatest for countries
with either or both large populations and high
per person GDP, including China, India, Japan
and the United States (Table 3.7 and Figure
3.15). It is important to emphasize that the WTP
does not reflect an ethical judgement about the
comparative value of the life of any individual,
but simply that the willingness to pay to reduce
risk is naturally a function of income. Hence
the larger valuation associated with lives lost in
wealthier nations merely reflects their wealth and
does not imply a greater value for their citizens.
Values in countries with lower per person
income are larger with the smaller value of
income elasticity, as is the global total for ozone-
related deaths which is approximately US$ 2500
(US$ 600–4 300; 95 per cent confidence interval
here and hereafter) per tonne of methane
emitted for an across country income elasticity
of 1.0 and US$ 5 300 (US$ 1 200–9 100) per
tonne for across country income elasticity of
0.4. Additional mortality-related benefits owing
to reduced heat-related deaths are US$ 110
(US$ 75–150) for impacts in the United States,
and US$ 120 (US$ 75–160) for China with income
elasticity of 1.0 and US$ 360 (US$ 230–490) for
China with income elasticity of 0.4. Using the
results from the national-average seasonal-mean
method to evaluate a broader set of countries
(Section 3.4.2), the valuation rises to US$ 700
(US$ 200–1 100) per tonne for an elasticity of
1.0 and US$ 1 400 (US$ 300–2 500) per tonne
for an elasticity of 0.4 and adopting the better
characterized uncertainty range associated with
ozone-related mortalities. These values are used
hereafter, noting again that they do not include
South Asia or Africa and are hence conservative.
The total worldwide mortality valuation of all
quantified impacts using an income elasticity of
1.0 is thus US$ 3 200 (US$ 800–5 400) per tonne
of methane.
As described in the previous section, several
additional impacts that can be readily monetized,
including health morbidity outcomes, labour
productivity in the United States, and crop yield
changes have been quantified. First, morbidity
changes associated with ozone exposure,
for which values are calculated for two health
impacts: hospital admissions and asthma-related
accident and emergency department visits. For
both end points, valuations are calculated only
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
for countries where baseline hospital admission
data for respiratory conditions are available
– China, Europe and the United States – or
accident and emergency department visits – a
broader set of countries.
First, the unit costs associated with each type
of morbidity burdens are estimated. However,
unit cost data is limited, as a result of which
the cost estimation involves data collection and
imputation, the latter for cases when actual data
are not available. In the data collection, one
important source was compiled by the CaRBonH
calculation tool (Spadaro 2018). It estimated unit
costs of several major morbidity endpoints for
53 countries, mostly in Europe. Since the target
pollutant for CaRBonH was PM instead of ozone,
the unit costs of morbidity endpoints studied were
not, however, immediately usable in this project.
CaRBonH reported country-specific average
costs of cardiovascular and respiratory hospital
admissions, for example, rather than reporting
respiratory hospital admission separately. In
addition, it reported the unit cost of each asthma
symptom day for those between the ages of
5–20, rather than costs for asthma accident and
emergency department visits for all age groups.
This analysis took advantage of the broad
coverage of CaRBonH’s country-level data but
made several adjustments to obtain estimates for
unit costs of ozone-related morbidity.
For hospital admissions, most countries for
which morbidity burdens could be estimated
are in Europe, due to the availability of baseline
hospitalization rates, except for China and the
United States and China. Thus the average unit
costs of hospital admissions for cardiovascular
(ICD-9 codes 390–459; ICD-10 codes I00–I99)
and respiratory (ICD-9 codes 460–519; ICD-10
codes J00–J99) concerns were collected for all
age groups from CaRBonH for all the European
countries in this analysis’ morbidity estimates,
and adjusted for respiratory issues as a single
cause, and for the post-65 age group. Detailed
cost information was used from the Healthcare
Cost and Utilization Project (HCUP) for the
United States to make the adjustment – the
HCUP’s dataset is the most comprehensive for
inpatients in the United States. The same ICD-9
codes from CaRBonH were used to locate the
diseases in HCUP dataset. The ratio of the mean
hospital costs for all (age) discharges between
cardiovascular and respiratory diseases, were
found to be 1.43, and the ratio of the mean
hospital costs for respiratory diseases between
those aged 65+ and all age groups was 0.98. The
unit costs for European countries were adjusted
using a scaling factor of 0.69 (= 0.98/1.43), to
75
obtain the estimated unit costs for respiratory
hospital admissions of those aged 65+. The
unit cost from HCUP for the United States was
used directly. To estimate the unit cost for China,
the average costs were calculated for three
respiratory diseases reported in China’s 2016
Health and Family Planning Statistical Yearbook,
weighted by the number of hospitalizations for
each disease. This estimate was not strictly
equivalent to all respiratory diseases considered
for other countries and was not specific to the
65+ population. Nonetheless, it is likely to be
a good approximation, since the three most
prevalent respiratory diseases in China were
selected, and the ratio between those aged 65+
and all age groups from HCUP, 0.98, indicated
that the hospitalization costs were likely to be
very similar between the two age-groups.
For the cost of visits to accident and emergency
departments for asthma, it was first note that
there is a highly linear relationship (R-square
= 0.96) between the cost of asthma symptom
days and of income per person across the 53
countries in CaRBonH. The cost of asthma
symptom days of children aged 5–19 analyzed
in CaRBonH is based upon health treatment
expenditures. Hence, here both the different age
groups were considered for the impact and the
difference between accident and emergency
department visits and general health treatment
for asthma. Since the cost for an accident and
emergency department visit is a proportion of all
treatment expenditure, it was assumed that there
was also a good correlation between the cost of
an accident and emergency department visit and
income per person worldwide. Secondly, it was
assumed that the correlation for children aged
5–19 holds for all age groups. Thus the cost
estimate utilized for each asthma accident and
emergency department visit for all age groups in
the United States, US$ 1 501 in 2008 US dollars
(Wang et al. 2014). As with mortality valuations,
income comparisons are made based on:
Costa = CostUS x (Incomea/IncomeUS)ß
In this case, the income elasticity (ß) follows
that used in CaRBonH, which is 0.8, 0.9, and
1.0 for high, upper-middle, and lower-middle
income countries, respectively. Income data
was obtained from the World Bank (World Bank
2020).
For respiratory hospital admissions, the largest
economic impact is in the United States, followed
by Germany, Italy, France and China within the
relatively limited set of countries (Table 3.8). The
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
total across all countries for which estimates were
made is US$ 0.36 per tonne. Values are generally
larger for the healthcare costs of asthma-related
accident and emergency department visits,
with highest expenditures in the United States,
Japan, India and China, respectively (Table 3.9;
Figure 3.16). The total across all countries for
which estimates were made is US$ 0.92 per
tonne. Hence, as expected, these healthcare
expenditures are comparatively small relative to
mortality-related valuations.
As previously discussed, labour losses due
to exposure to extreme heat were calculated
using two distinct methodologies. The first
was applied only for the United States and,
using this method, a 50 per cent reduction in
anthropogenic methane concentrations led
to a reduction in lost labour of 13.6 million
work hours. To estimate the value associated
with the time lost, 2016 county-level annual
employment for those of working ages, 15–64
year-olds, and annual average weekly wages
from the Quarterly Census of Employment and
Wages (U.S. Bureau of Labor Statistics 2020a;
accessed 10 January 2019) were used. The
following North American Industry Classification
System (NAICS) codes were used to determine
the number of workers in high-risk industries/
sectors: NAICS 11 for agriculture, forestry,
fishing and hunting; NAICS 23 for construction;
NAICS 31-33 for manufacturing; NAICS 21 for
mining; NAICS 48–49 for transport, and NAICS
22 for utilities (U.S. Bureau of Labor Statistics
2020b; accessed 10 January 2019). The 2016
fraction of workers in high-risk industries was
used for each county. National average wages
range from US$ 19.0 per hour for agriculture to
US$ 44.90 per hour for utilities and are spatially
heterogeneous across the United States.
Accounting for the affected economic sectors,
largely manufacturing and construction, these
losses are valued at US$ 449 (73–601; 95 per
cent confidence interval) million. At the state
level, the largest impacts are in Texas, US$ 141
million; Arizona, US$ 32 million; and Florida,
US$ 24 million, with impacts generally larger in
warmer states.
The second method for estimating of labour
losses due to extreme heat was applied globally.
Hours of reduced work availability calculated
using this method were valued based on hourly
wage data for the affected work areas from the
ILO (ILO 2020; accessed 2 September 2020).
This dataset provided enough detail to attribute
wages to the three labour categories described
previously (light, medium and heavy) for 131
76
countries. The hourly wage data provided by
ILO varies from country to country, ranging from
2014 to 2018, and were converted to US dollars
based on national purchasing power parity. For the
United States, the latest year, 2015, as reported
by ILO was used, adjusted to 2018 US dollars.
For the United States, the labour loss reductions
from a 50 per cent decrease in anthropogenic
methane concentrations were 44.1 million hours.
These losses are valued at US$ 531 million per
year (360–702; 95 per cent confidence interval),
similar to the results using the first method. This
similarity, however, is largely coincidental as
the labour losses are much larger in the second
method but that is offset by markedly lower hourly
wages across all sectors in the United States in
the ILO dataset relative to government statistics.
There are also differences in the employment
sectors affected. The strongest effects in the first
method are in the manufacturing, construction and
transport sectors. In the second method, there are
again significant impacts in construction which
is considered part of the heavy outdoor work
category, the category that represents the vast
majority of impacts, but there are no impacts on
manufacturing or transport due to the assumption
of a lower WBGT for these sectors as they are
considered indoor medium work. Hence the labour
loss results are quite sensitive to the wage dataset
used and the methodology employed.
The global total using this second method was
a value of US$ 8.5 billion (3.1–13.8; 95 per cent
confidence interval) for the half anthropogenic
methane reduction, corresponding to value of
US$ 63 per tonne of methane (US$ 23–103 per
tonne). Effects are large in both hot and humid
countries where the hours lost are greatest such
as Brazil, China, India and Indonesia, but also in
some less sensitive but wealthier countries such
as Australia, the United States and the United
Arab Emirates (Table 3.10; Figure 3.17). The total
is substantially larger than valuations associated
with ozone-related morbidity outcomes across
all countries with estimates of accident and
emergency department visits for asthma and
hospital admissions but is similarly much less than
the value associated with ozone-related or heat-
related mortalities. There would be differences
in the valuation of the lost labour for men and
women, as was seen for the labour hours lost
(Table 3.5b). These are not quantified here as the
ILO dataset underlying these calculations does
not include wage differences by gender.
For changes in crop production, valuation is
based upon yield changes multiplied by the
2018 global market prices per tonne from the
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
United States Department of Agriculture (USDA)
and World Bank11 which gives US$ 210 for
wheat, US$ 394 for soybeans, US$ 164 for
maize and US$ 421 for rice. This provides a fairly
simplistic measure of the value of yield changes,
as it clearly does not account for benefits such
as those to subsistence farmers or national food
security. The valuation per tonne of methane is
then US$ 11.7 for wheat, US$ 6.6 for soybeans,
US$ 6.8 for maize and US$ 13.2 for rice leading
to a valuation summing over these four crops of
just more than US$ 38 per tonne of methane. For
comparison, in prior studies that only accounted
for the effects of ozone on yields, the value
was US$ 29 per tonne of methane (Shindell
et al. 2012; UNEP 2011). As climate impacts,
predominantly from temperature change, are
now included it seems reasonable that the
new values is roughly one third larger. While
at the global level valuation of reduced risk of
premature death dominates all other impacts,
crop changes can be of comparable magnitude
in some areas depending on the income elasticity
used to account for differential income levels
across countries. Using an elasticity of 1.0, for
example, leads to agricultural impacts in India
having a larger valuation than that of the reduced
risk of premature death, and a valuation for
Brazil that is more than half that of the reduced
risk of premature death (Figures 3.15 and 3.18).
It should be note that there are additional
potential effects from air quality that were not
included here, such as changes in forestry
yields, tourism or depreciation of man-made
materials. The valuation of the impacts of ozone
resulting from methane emissions on forestry
have been evaluated as ~US$ 20 per tonne of
methane, based on a global extrapolation of a
United States analysis (West and Fiore 2005).
Finally, the effects of climate change using a
social cost framework was assessed. Social
costs of climate change attempt to account for all
the effects of climate, including both market and
non-market costs, but ignore the effects of air
pollution. This makes it challenging to compare
with the values in this analysis as the agricultural
impacts included, for example, are the result of
both climate and air pollution changes. There
was a clear rationale for excluding air pollution
in the original development of social costs for
carbon dioxide given that emissions of carbon
dioxide and co-emitted air pollutants are not
necessarily correlated. For example, two similar
coal-fired power plants or motor vehicles (with
internal combustion engines) could have quite
11. indexmundi.com/commodities
77
different air pollutant emissions depending on
the efficiency of any applied pollution controls
such as scrubbers or catalytic converters. Hence
conventional practice has been to include the
response of air pollutants such as ozone to carbon
dioxide emissions when this takes place due to
climate change but to exclude the response to
co-emissions that takes place via atmospheric
chemistry. This has extended not only to social
costs, but to broader evaluations of the societal
impacts of greenhouse gas emissions such as
the US EPA’s Climate Change Impacts and Risk
Analysis project (US EPA 2015a). With methane,
however, this rationale breaks down as methane
affects ozone both via climate change and
atmospheric chemistry. Following conventional
practice and leaving out atmospheric chemical
responses therefore means that an intrinsic
effect of methane emissions themselves is
neglected rather than an effect of a potentially
co-emitted pollutant.
Prior analyses of the climate impacts of methane
yield a per tonne social cost of US$ 810 (Marten
and Newbold 2012) or US$ 910 (Shindell 2015) in
2007 US dollars and using a 3 per cent economic
discount rate. Values without discounting, as in
the VSL analyses here, would of course be larger
but are not standard in the existing literature. The
largest climate-related impact in the analysis
was the finding that the reduced risk of heat-
related premature mortalities in China and the
United States alone were valued at ~US$ 230
per tonne, and that a rough estimate of the
worldwide valuation was US$ 700 per tonne.
Applying 3 per cent discounting to that estimate
would reduce the US$ 700 per tonne to US$ 580
per tonne. This analysis therefore suggests that
prior estimates of the climate-only social cost of
methane are potentially too low. As such, this
analysis includes the valuation of heat-related
health impacts along with these social costs
as they account for damage associated with
all other impacts and appear to underestimate
the effects of climate change on human health
(Carleton et al. 2020; Shindell 2015). Here, the
mean of the studies mentioned above is used,
which is US$ 1 100 per tonne converted to 2018
US dollars and rounded to the nearest hundred.
Note that values for fossil methane are slightly
higher, by ~US$ 150 per tonne with a 3 per
cent discount rate (Shindell et al. 2017), due to
the climate impacts of the carbon dioxide that
is eventually produced by the oxidation of the
carbon within the emitted methane.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES
The total valuation per tonne of methane for
all market and non-market impacts assessed
here is roughly US$ 4 300 using a cross-nation
income elasticity for WTP of 1.0 and US$ 7 900
using an elasticity of 0.4 (Figure 3.19) – values
are ~US$ 150 per tonne larger for fossil-related
emissions. This value is dominated by mortality
effects, of which US$  2  500 are due to ozone
and ~US$ 700 are due to heat using the more
conservative 500 deaths per million tonnes
of methane of this analysis’ two global-scale
estimates and a WTP income elasticity of 1.0,
followed by climate impacts. These numbers
are comparable to those reported previously
using a social cost framework that incorporated
both climate and air pollution related impacts
(Shindell et al. 2017), though slightly larger as
a result of the updated epidemiology underlying
the estimates of ozone-attributable premature
deaths. Given the dominance of ozone-related
impacts that occur immediately in response
to methane changes, decaying thereafter with
the residence time of methane, these impacts
are only weakly sensitive to discounting. For
example, the ozone-related impacts over the first
25 years following methane emissions, using 5
per cent annual discounting, are 83 per cent of
the value using 2 per cent discounting and the
value is 80 per cent over the first 50 years. In
contrast, the valuation of a specific impact 25
years in the future using 5 per cent discounting
is only 50 per cent of the value using 2 per cent
discounting, a proportionality more applicable to
the effect of temperature change based on the
evaluation of the response averaged over 10–40
years after emissions. Hence roughly two thirds
of the total valuation has minimal sensitivity to
discounting whereas the remainder would be
substantially affected were future benefits to be
discounted at a rate other than the 3 per cent
assumed here for climate, which is the dominant
contributor to the third of impacts sensitive to
discounting. Using a discount rate of 5 per cent,
for example, the value of benefits would be
~US$ 700 less whereas using a discount rate of
1.4 per cent it would be ~US$ 700 greater.
The total market costs are comparatively small
at ~US$ 122 per tonne and stem predominantly
from labour losses, agricultural yield changes
and forestry losses. In contrast, healthcare costs
related to morbidity effects associated with
ozone were relatively small, about US$ 1 per
tonne, though there are likely to be additional
78
non-fatal health outcomes associated with
ozone. It was noted that although discounting is
conventionally applied to estimates of mitigation
costs (Section 5), the valuation of impacts does
not discount future benefits. This is consistent
with the usage of impact valuations to assess
effects on future generations, especially in the
case of climate change related impacts that
occur over many decades, and the desire to
value welfare of future generations similarly
to that of today’s population, which implies a
near-zero discount rate. This is in contrast to
the use of mitigation costs to compare options
for the use of capital that employ a variety of
discount rates that are not near zero. This
issue is discussed further in Section 5. As the
valuation of benefits is dominated by the effects
of air pollution, and this responds immediately
to changes in methane emissions, with impacts
tapering off according to the ~12-year methane
residence time, these benefits are in any case
much less sensitive to the choice of discounting
rate than are climate-related benefits.
A prior valuation of the impacts of methane
through ozone on agriculture, forestry, and
non-fatal health endpoints found a monetized
impact of ~US$ 80 per tonne (West and Fiore
2005). This resulted from estimated impacts
of roughly US$ 30 per tonne due to both
agricultural losses and non-fatal human health
effects, along with the ~US$ 20 per tonne due to
forestry losses. The agricultural impacts shown
in this analysis are larger, consistent with the
prior study not including climate impacts – it
should also be noted that West and Fiore (2005)
also extrapolated globally from limited regional
studies in the United States, Europe and East
Asia. Their valuation of non-mortality human
health effects is markedly greater, however.
They again extrapolated from limited studies,
in this case in the United States and Europe
only, but more importantly included different
impacts, namely avoided minor restricted
activity days and worker productivity changes
in the United States and morbidity impacts
in Europe associated with ozone exposure,
for both of which there is some, but limited,
evidence. Hence the results here are larger for
market valuations because this analysis found
larger impacts for crops and included impacts
on labour, which had a larger influence than not
including several of the impacts that contributed
most strongly in West and Fiore’s 2005 analysis.
Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 79

Table 3.7a Value of reduced risk of ozone-related deaths per tonne of methane for countries with values greater
US$ 10*

COUNTRY US$ (MILLION) LOW HIGH

United States 639 159 1 077

China 390 95 653

Japan 210 57 346

Germany 161 38 278

United Kingdom 133 32 229

India 85 13 149

Italy 74 18 126

Russia 74 18 127

France 72 16 124

Spain 59 14 100

Canada 39 10 67

Netherlands 32 7 56

Brazil 30 6 53

Australia 27 6 48

Korea, Rep. 27 7 45

Belgium 23 5 40

Poland 21 5 37

Indonesia 19 3 35

Mexico 18 4 31

Switzerland 17 4 29

Sweden 17 4 30

Turkey 15 4 25

Argentina 14 3 25

Saudi Arabia 14 4 23

Denmark 14 3 23

Norway 13 3 22

Austria 13 3 22

Portugal 12 3 21

Romania 12 3 21

Greece 11 3 18

Ireland 10 2 17

Czechia 10 2 17
GLOBAL 2 511 598 4 264

Note: income elasticity of 1.0 across countries

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 80

Table 3.7b Value of reduced risk of ozone-related deaths per tonne of methane for countries with values greater than
US$ 10*
US$ US$
COUNTRY LOW HIGH COUNTRY LOW HIGH
(MILLION) (MILLION)

China 1 189 291 1 994 Belgium 27 6 47

India 671 100 1 179 Saudi Arabia 25 6 42

United States 639 159 1 077 South Africa 25 6 43

Japan 278 75 459 Portugal 22 5 38

Russia 206 49 354 Morocco and

22 5 37
Western Sahara
Germany 191 45 328
Greece 21 5 35
United Kingdom 168 40 288
Sweden 19 4 32
Italy 107 26 180
Czechia 18 4 31
Indonesia 103 14 185
Kazakhstan 18 4 30
Brazil 98 21 172
Malaysia 17 3 31
France 92 21 159
Myanmar 17 4 29
Spain 91 22 154
Algeria 16 4 28
Pakistan 72 14 124
Uzbekistan 16 4 26
Egypt 64 17 106
Hungary 16 4 27
Ukraine 57 13 98
Nigeria 15 2 27
Mexico 56 13 94
Switzerland 15 4 25
Poland 49 11 85
Austria 14 3 24
Canada 47 12 80
Denmark 14 3 24
Turkey 47 12 79
Serbia and
14 3 23
Korea, Rep. 41 11 68 Montenegro

Cuba 14 3 24
Philippines 41 9 72
Bulgaria 13 3 22
Argentina 39 9 69
Sri Lanka 13 2 23
Iran, Islamic Rep. 36 10 60
Chile 12 3 21
Netherlands 36 8 62
Ethiopia 12 2 22
Bangladesh 35 7 59
Peru 12 2 22
Romania 32 7 55
Belarus 11 3 20
Viet Nam 30 6 51
Norway 11 3 19
Australia 29 6 51
Israel 11 2 18
Thailand 28 5 50
GLOBAL 5 315 1 193 9 073

Note: income elasticity of 0.4 across countries

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 81

Table 3.8 Country-specific cost of respiratory hospital admissions attributable to ozone exposure, 2018 US$ per
‘000 tonnes, 65+ age group

COUNTRY COST (2018 US$)

United States 83

Germany 54

Italy 46

France 37

China 31

United Kingdom 26

Spain 25

Belgium 6.2

Greece 5.5

Netherlands 5.4

Austria 5.3

Poland 5.2

Norway 4.9

Sweden 4.8

Portugal 3.6

Denmark 3.1

Hungary 1.6

Czechia 1.5

Ireland 1.4

Finland 1.3

Bulgaria 1.2

Romania 1.1

Slovakia 1.0

Slovenia 1.0

Croatia 0.6

Serbia and Montenegro 0.4

Luxembourg 0.3

Lithuania 0.3

Estonia 0.2

Cyprus 0.2

Malta 0.2

Iceland 0.1

Latvia 0.05
TOTAL 358

The uncertainty range is ± 84 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 82

Table 3.9 Cost of asthma-related accident and emergency department visits attributable to ozone exposure, US$ per
‘000 tonnes for countries with values greater than US$ 5 per ‘000 tones, all age groups

COUNTRY US$ PER ‘000 TONNES

United States 278

Japan 78

India 66

China 52

Spain 49

Korea, Rep. 41

Norway 39

Canada 33

Germany 26

Netherlands 24

Brazil 22

United Kingdom 19

Turkey 18

Australia 15

Sweden 14

United Arab Emirates 13

Russian 13

Italy 12

France 11

Argentina 8.6

Belgium 8.4

Kuwait 5.6

Venezuela 5.5

Colombia 5.1

Austria 5.0

TOTAL 924

The uncertainty range is ± 37 per cent.

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 83

Table 3.10 Value of labour losses due to heat exposure, 2018 US$, top 15 countries

COUNTRY COST PER TONNE (2018 US$)

China 32.7
Brazil 5.3
Australia 4.7
United States 3.7
India 2.4
Indonesia 1.7
Viet Nam 1.4
United Arab Emirates 1.1
Philippines 1.0
Thailand 0.8
Qatar 0.6
Malaysia 0.5
Mexico 0.5
Pakistan 0.5
Saudi Arabia 0.5
Global 63

The uncertainty range is ± 62 per cent.

Figure 3.15 Valuation of reduced risk of premature death due to respiratory and cardiovascular illnesses caused by
ozone, people aged 30+ per million tonnes of methane emission, US$ millions

Note: Grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 84

Figure 3.16 Valuation of the change in asthma-related accident and emergency department visits due to ozone
exposure change per million tonne increase in methane emissions, 2018 US$ ‘000

Note: Based on the scaled multi-model mean ozone change. Grey indicates no quantification.

Source: UNEP and CCAC

Figure 3.17 Valuation of labour productivity gains due to reduced heat exposure per million tonne reduction in
methane emissions, 2018 US$ ’000

Note: Grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / COMPOSITION-CLIMATE MODELLING AND IMPACT ANALYSES 85

Figure 3.18 Valuation of the avoided yield losses of soybeans rice, wheat and maize per million tonne reduction in
methane emissions, 2018 US$ ‘000

Note: Grey indicates no quantification.

Source: UNEP and CCAC

9 000 Figure 3.19. Valuation

per tonne of emitted
methane apportioned
8 000
to the various impacts
quantified in this
SOCIAL COST PER TONNE OF METHANE (2018 US$)

7 000 assessment for the

indicated elasticity in
cross-nation income
6 000 comparisons, 2018 US$
Note: Heat VSL and
5 000 climate impacts are
strongly sensitive to the
4 000
assumed discount rate.
Market costs are
dominated by
3 000 agricultural and labour
impacts, which are
2 000 moderately sensitive to
discounting.
1 000 Source: UNEP and
CCAC
0
INCOME ELASTICITY = 1.0 INCOME ELASTICITY = 0.4

MARKET COSTS HEAT VSL CLIMATE OZONE VSL

4. METHANE
EMISSIONS
MITIGATION
CHAPTER FINDINGS
• In the absence of additional policies, methane emissions are projected
to continue to rise through at least 2040 with emissions at the end of the
century higher than at present in all but one socio-economic pathway.

• In modelled scenarios consistent with keeping global mean temperatures

below the 1.5ºC target, methane emissions are reduced in response to
targeted and additional measures by around 100–150 Mt/yr from the fossil fuel
and waste sectors combined and by about 30–80 Mt/yr from the agricultural
sector in 2030 relative to reference case emissions, with a substantial range
across models and across baseline socio-economic pathways.

• A
 nalysis of the technical potential to mitigate methane from four separate
studies shows that for 2030, reductions of 29–57 Mt/yr could be made in
the oil and gas subsector, 12–25 Mt/yr from coal mining, 29–36 Mt/yr in the
waste sector and 6–9 Mt/yr from rice cultivation. Values for the livestock
subsector are less consistent, ranging from 4–42 Mt/yr. Average cost
estimates vary greatly across the analyses, not only in amounts but even
as to whether they are net expenditures or savings for some sectors.

• T
 argeted abatement measures with a negative cost could reduce total
emissions by ~40 Mt/yr with the greatest potential for abatement coming
from the oil and gas subsector and the waste sector. Considering only
low-cost targeted controls, defined here as less than US$ 600 per tonne
of methane, abatement potentials still typically represent a substantial
fraction of the total potential, from 60–80 per cent for oil and gas in three
of the four analyses, but 36 per cent in the fourth which does not consider
negative costs, from 55–98 per cent for coal, and from ~30–60 per cent in
the waste sector.

• C
 onsidering the potential for mitigation in different sectors and regions, the
largest potential in Europe and India is in the waste sector, in China from
coal production followed by livestock, in Africa from livestock followed by
oil and gas, in the Asia-Pacific region, excluding China and India, from the
coal subsector and the waste sector, in the Middle East, North America
and Former Soviet Union it is from the oil and gas subsector, and in Latin
America from the livestock subsector. A majority of these major abatement
potentials can be achieved at low cost, less than US$  600 per tonne of
methane, especially in the waste sector and coal subsector in most regions
and for the oil and gas subsector in North America.

• W
 ithin the oil and gas subsector, across available analyses mitigation
potentials are greatest for oil production, followed by natural gas production
and then downstream gas. With regard to reducing emissions from the
waste sector, most of the potential is from changes in the treatment and
disposal of solid waste.
• A
 chieving climate stabilization at levels that avoid dangerous anthropogenic
interference with the climate requires the decarbonizing of economies,
which will affect methane emissions. Decarbonization of the economy
results in only about 30 per cent of the methane abatement seen under
a broad multi-pollutant, multi-policy 2ºC scenario in 2050, however,
highlighting the role of methane-specific policies.

• B
 y 2050, emissions from enteric fermentation, especially from cattle, are
by far the dominant remaining source of methane emissions in under 2ºC
scenarios produced by IAMs as emissions from other sources have been
substantially reduced.

• T
 here are numerous financial, legislative and regulatory mechanisms
available to support implementation of the identified targeted and additional
measures at scale. One commonly employed in IAMs is the imposition of
a fairly modest methane tax, which could lead to rapid and substantial
decreases in methane emissions, especially from coal, gas and landfills,
with abatement of livestock-related emissions highly model-dependent.

• G
 iven the limited technical potential to address agricultural sector methane
emissions, behavioural change and policy innovation are particularly
important for this sector.

• A
 relatively robust evidence base indicates that three behavioural changes,
reduced food waste and loss, improved livestock management, and
adoption of healthier diets, have the potential to reduce methane emissions
by 65–80 Mt/yr over the next few decades.

• A
 web-based decision support tool has been created that allows users to
input a level of methane emissions mitigation or select from the analyzed
abatement options and view the national and global values for all impacts
included in this assessment.

As discussed in Chapter 3, the modelling performed for this assessment

quantifies many physical and societal impacts associated with potential
changes in methane concentrations. Those changes are evaluated in terms of
the corresponding change in emissions using the well-established relationship
between methane emissions and concentrations, and the analysis presented in
Chapter 3 shows that the impacts are generally well-represented by assuming
a linear relationship between emissions changes and impact responses. This
chapter examines changes in methane emissions both in scenarios consistent
with low warming targets and in response to specific methane reduction policies
or technologies. The impacts analyzed in Chapter 3 can then be related to the
methane abatement potentials explored here, as is primarily done in Chapter 5.
Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.1 METHANE EMISSIONS UNDER
BASELINE AND ALL-SECTOR ALL-
POLLUTANT MITIGATION SCENARIOS
There are a multitude of internally consistent
projections of future methane emissions.
Among those, the SSPs that are the result of a
coordinated effort by the research community to
generate consistent future scenarios that span a
broad range of possible societal developments
with a variety of IAMs are emphasized
(Riahi et al. 2016). The underlying storylines
include developments in socioeconomics,
demographics, technology, lifestyle, policy
and institutions. They consist of five distinct
narratives, referred to as sustainability (SSP1),
middle-of-the-road (SSP2), regional rivalry
(SSP3), inequality (SSP4), and fossil-fueled
development (SSP5). These scenarios are
intended to span a range of challenge levels for
both mitigation and adaptation. For mitigation,
challenges are rated as relatively small for SSP1
and SSP4, medium for SSP2, and high for SSP3
and SSP5. Each scenario includes a baseline
case as well as scenarios with increasingly
stringent climate change mitigation goals. To
examine the question of how much methane
mitigation is achievable and how much is
consistent with worldwide targets for climate
change mitigation, this analysis is particularly
interested in scenarios consistent with the 1.5ºC
target as assessed in the 2018 IPCC Special
Report on Global Warming of 1.5ºC (IPCC
2018; Rogelj et al. 2018). Here those scenarios
classified by IPCC as either having at least a
50 per cent likelihood of keeping temperatures
below 1.5ºC throughout the 21st century and
those with a median temperature below 1.5ºC
in 2100 but with a 50–66 per cent change of
overshooting that value during the century,
typically by no more than 0.1ºC, are considered.
Many, but not all, of these scenarios were
generated within the SSP framework.
Given the wide range of socio-economic
assumptions, it is not surprising that projected
methane emissions over the remainder of this
century vary markedly across the baseline
cases for the SSPs (Figure 4.1). Through 2060,
emissions are greatest for SSP3 and SSP5, but
from 2070 onward they are largest in SSP3 as
emissions in some regions continue to grow
rapidly throughout the century. This increase is
largely driven by agriculture as food demands
increase with an ever-growing population that
exceeds 12 billion by 2100. Under SSP3, no
models were able to produce a 1.5ºC scenario.
Models were able to generate 1.5ºC and 2ºC
89
scenarios for the other four SSPs, though not all
models could. Given the difference in baseline
emissions across those four SSPs, evaluation
of methane mitigation under the 1.5ºC or 2ºC
scenarios as a fraction of projected emissions
therefore depends greatly on the reference
pathway. Even for 2030, there are substantial
differences across the baselines with SSP4 and
SSP5 having emissions of about 490–500 Mt/
yr whereas SSP1 and SSP2 have emissions
of about 390 Mt/yr. It should be noted that
these also diverge by around 40 Mt/yr in 2020,
reflecting uncertainties in current methane
sources, as discussed in Chapter 2, and that the
divergence is substantially larger in 2030.
Most 1.5ºC scenarios are in agreement with
recent observations for methane amounts
(Figure 4.2). This is in contrast to the older
representative concentration pathways (RCPs)
that do not capture recent trends (Nisbet et al.
2020), with the better performance of the SSPs
unsurprising as they were created later when
these observations were known. This analysis
has screened out the few scenarios that do
not agree with observations made in the past
decade for methane. The scenarios included in
this assessment come from seven distinct IAMs,
though not all provided all diagnostic outputs.
The 1.5ºC scenarios show marked declines
in methane amounts over the coming 20
years (Figure 4.2). These are achieved by
lowering emissions in the current decade from
the agriculture, forestry and other land-use
category, the waste sector and especially the
energy sector (Figure 4.3). All decrease energy-
and waste-related emissions and all but one
decrease emissions from agriculture, forestry
and other land-use. In all three sectors there are
large variation across models and scenarios.
For the agriculture, forestry and other land-use,
the average reduction in emissions in 2030 is 22
per cent compared with 2020, with a standard
deviation of 18 per cent across models and a full
range of 0–51 per cent. For the energy sector,
the average reduction of emissions in 2030
is much larger at 59 per cent relative to 2020,
with a smaller standard deviation of 14 per cent
across models and a full range of -29 per cent to
-82 per cent. For the waste sector, the average
reduction of emissions in 2030 is between the
other two at 34 per cent relative to 2020, with
a substantial standard deviation of 23 per cent
across models and a full range of -6 per cent
to -78 per cent. The average reduction in the
total from these three sectors is 37 per cent
reduction in anthropogenic methane emissions
in 2030 relative to 2020. Hence the 1.5ºC
Global Methane Assessment / METHANE EMISSIONS MITIGATION
scenarios provide a wide range of estimates
of potential methane mitigation relative to the
present. Assuming the relative mitigation across
subsectors is similar to that reported for most of
these same IAMs under 2ºC scenarios in 2050,
this translates to average 2030 reductions of
about 20 per cent for enteric fermentation, 22
per cent for manure management and 30 per
cent for rice cultivation. For energy and waste,
estimated reductions are about 75 per cent for
coal mining related emissions, 60 per cent for oil
and natural gas, 65 per cent for landfills and 45
per cent for sewage.
Analysis of the SSPs, which include some 1.5ºC
scenarios but not the full dataset shown in Figures
4.2 and 4.3, provides estimates of mitigation
potential in a given future year relative to the
reference case emissions for that year. Despite
reference emissions varying greatly across the
SSPs, mitigation in 2030 is fairly consistent
across them (Figure 4.4). Examining the mean
across IAMs, excepting SSP4, for which there is
only one model that produced a 1.5ºC scenario
and that model’s results also do not vary greatly
across SSPs, methane abatement is around
50–70 Mt/yr- for land-use, primarily agriculture,
and around 120–130 Mt/yr for the energy and
waste sectors combined. There is a large range
of results across the different IAMs, however,
which show mitigation levels that differ by up to
a factor of about seven in land-use and about
two and a half in the energy and waste sectors
combined. For SSP1 and SSP2, for which at least
four models produced 1.5ºC scenarios, the range
for land-use mitigation in 2030 is in the range of
17–125 Mt/yr whereas for the energy and waste
sectors combined the range is 69–192 Mt/yr.
Fuel switching plays an important role in the
decreases in methane emissions from the energy
sector in most of the IAMs. Transitioning to
renewables, for example, would in the long term
remove the bulk of methane emissions, and is
what happens in most IAMs in order to achieve
stringent mitigation targets. This does not happen,
however, in all pathways consistent with low
warming targets. The scenarios examined in 2018
IPCC Special Report on Global Warming of 1.5ºC
and classified as having no or limited overshoot,
for example, found the 25th percentile showed a
decrease of 26 per cent in primary energy from
gas whereas the 75th percentile suggested an
increase of 21 per cent in 2030 relative to 2010
(IPCC 2018). The full range across those scenarios
shows a median decrease of 15 per cent in
primary energy from gas between 2020 and 2030,
but the range is extremely broad, expanding from
105–153 exajoules (EJ) in 2020 to 17–174 EJ in
90
2030 (Rogelj et al. 2018). As a result, there is a
possibility that the use of gas, and hence fugitive
methane emissions from the gas subsector as
well as emissions associated with its extraction,
will increase even under aggressive climate
change mitigation policies. This is quite distinct
from the use of coal or oil, which drop much more
uniformly across all potential scenarios. Most of
the scenarios that include increases in gas use
over the next decades maintain consistency
with the stringent climate targets by relying on
large-scale negative-emissions technologies
in the future to compensate for their high early
emissions. This is primarily biofuel energy with
carbon capture and storage (BECCS), which
does not currently exist at scale, is currently
quite expensive though the models project that
it might be inexpensive many decades into the
future, and requires converting large areas of land
to growing biofuel crops rather than food with
associated issues around food security, water
usage, fertilizer application and biodiversity,
which has led to popular protests against even
the limited current deployment of biofuel plants
without carbon capture and storage. Biofuel
energy with carbon capture and storage is also
not carbon negative in the short-term and can in
fact lead to temporarily increased carbon, when,
for example, a forest is cut down for biofuel use
(Sterman et al. 2018; Holtsmark 2012). Many of
these barriers also exist for the carbon capture
and storage portion alone, as it similarly does not
currently exist at scale and remains expensive.
Hence there are multiple risks that this technology
will not work, will be too expensive, and/or will
have so many side effects that society will not
want to use it. Those scenarios also generally
overshoot the 1.5ºC target substantially, only
returning to an appropriate level towards the very
end of this century.
Scenarios that do not overshoot the target or rely
heavily on biofuel energy with carbon capture
and storage generally feature a shift away from
gas to renewables, but in a small number of those
use of gas goes up over the current decade as
well. In those pathways, afforestation can be
very large and carbon capture and storage is
eventually applied to natural gas plants, turning
them in to approximately carbon neutral power
sources. This technology does not raise the
issues of water and land use associated with
biofuel energy with carbon capture and storage,
but does still assume that carbon capture and
storage technology drops dramatically in price
so that it becomes fairly inexpensive to have
gas with carbon capture and storage before
mid-century. Thus, there is still a range from
 Global Methane Assessment / METHANE EMISSIONS MITIGATION 91

steep declines in gas usage for primary energy
to substantial increases through 2030, though
the typical values across those scenarios are
decreases of ~15–30 per cent, with the largest
decreases being about 85 per cent. When the
analysis is restricted to only those scenarios
in which the average 2040-2060 BECCS and
afforestation values are fairly low (less than 5.0
Gt CO2/yr and 3.6 Gt CO2/yr, respectively), global
production of gas has to decline annually by
~3% over 2020-2030 to be consistent with a 1.5°
C pathway (SEI et al. 2020). This corresponds to
a decrease in gas usage by roughly one-third by
2030, whereas current plans and projections are
for an increase of ~20 per cent relative to 2020
usage (SEI et al. 2020).
It is worth emphasizing the dramatic changes
required in scenarios that do not rely on large-
scale future deployment of carbon capture and
storage, whether for bio or fossil fuels. Analyses
show that without large amounts of carbon
removal, given the large reduction in gas usage
required under low warming pathways even the
current fossil fuel infrastructure is incompatible
with the world’s climate targets, with the situation
much worse when planned additional fossil fuel
energy sector infrastructure is also considered
(Shearer et al. 2020; Pfeiffer et al. 2018). This
implies that use of current power plants would
have to be curtailed either by early retirement
or a reduction in usage and construction of
additional fossil capacity halted in the absence
of large-scale carbon removal in order to meet
the well-below 2ºC target.
In addition, a transition away from fossil fuels
could still leave abandoned infrastructure. There
were more than 3.2 million abandoned oil and
gas wells in the United States alone in 2018,
which emit ~0.3 Mt/yr of methane according to
the US EPA (US EPA report to UNFCCC; 2020).
That agency acknowledges that this figure is
likely a large underestimate due to incomplete
data. Similarly, The International Institute for
Applied Systems Analysis (IIASA) estimates that
2020 emissions of methane from abandoned
coal mines around the world are just over 3.5
Mt/yr (Höglund-Isaksson 2020).

900

800

700
METHANE EMISSIONS Mt/yr

600

500

400

300

200
•

100

0
2020 2030 2040 2050 2060 2070 2080 2090 2100

AIM/CGE SSP3-BASELINE GCAM4 SSP4-BASELINE

IMAGE SSP1-BASELINE MESSAGE-GLOBIOM SSP2-BASELINE
REMIND-MAGPIE SSP5-BASELINE

Figure 4.1 Estimated global total anthropogenic methane emissions under the five shared socioeconomic pathway
baseline scenarios, 2020–2100, million tonnes per year

Values are shown for the marker scenario with a particular integrated assessment model’s results
chosen as a representative example for each shared socioeconomic pathway.

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION 92

Figure 4.2 Estimated global average methane amounts under scenarios consistent with 1.5ºC, and as observed for
2010–2020, parts per billion

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION 93

Figure 4.3 Estimated total global anthropogenic methane emissions under scenarios consistent with 1.5ºC from the
land-use (top), energy (middle) and waste (bottom) sectors, 2010–2100, million tonnes per year

Note: All available scenarios that included all three sectors are shown.

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION 94

AGRICULTURE, FORESTRY AND OTHER LAND USE

METHANE (Mt/yr)

ENERGY AND WASTE

METHANE (Mt/yr)

Figure 4.4 Estimates of methane mitigation potential for the land-use (top) and the energy and waste (bottom)
sectors under the four shared socioeconomic pathways for which at least one model produced a 1.5ºC scenario,
2030, million tonnes of methane per year

Note: Bars show the mean across integrated assessment models and ranges indicate maximum
and minimum values – there is no range for SSP4 as only one produced a 1.5ºC scenario under that
pathway. Similarly, the range is very small for SSP5 as only two models produced a 1.5º C scenario.
Agriculture, forestry and other land use includes biomass burning. Energy and Waste includes small
contributions from biofuels and transportation.

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.2 MARGINAL ABATEMENT
POTENTIALS AND COSTS BY SECTOR
The methane abatement that takes place in the
all-sector all-pollutant climate change mitigation
scenarios discussed in Section 4.1 is based
in part on abatement cost curves that are
incorporated within the IAMs. These encompass
methane mitigation opportunities in multiple
sectors at a variety of costs, which are examined
in detail in this section. Abatement potentials
have been analyzed by several teams, including
IIASA, US EPA, the International Energy Agency
(IEA) and in a literature review by Harmsen et
al. (2019). These analyses are similar in their
aims but differ in coverage and methodology.
In particular, the IEA analysis includes only the
oil and gas subsector and analyzes present-day
abatement potentials associated with targeted
control measures (IEA 2020). In contrast, the
IIASA analysis covers all sectors and include
time-dependent estimates of both changes
in baseline emissions and mitigation in which
the latter include sector-specific assumptions
about technology turnover times, based on the
literature, improvements in technology over
time and the achievable pace of regulations
(Höglund-Isaksson 2020). The IIASA analysis,
furthermore, includes discount rates of 4 and
10 per cent in their cost evaluation and extends
to 2050. The US EPA produced analyses in
2019 that similarly include projected changes in
baseline emissions, extend to 2050, and use a
discount rate of 10 per cent in cost estimates
(US EPA 2019). Finally, the Harmsen et al. (2019)
analysis covers all sectors, includes estimates
of technology development and the removal
of implementation barriers through 2050 as
implemented in the Integrated Model to Assess
the Global Environment (IMAGE) IAM following
SSP2, and uses a discount rate of 5 per cent.
This analysis is not entirely independent of
the others as it relies upon IIASA data for the
fossil fuel sector and earlier US EPA data (US
EPA 2013) for the waste sector but implements
different projections of future development and
so obtains different results even in those sectors
as well as in the agricultural sector for which
it provides an independent analysis. Note that
the IEA documentation does not describe the
discount rate used in its analysis.
To create compatible regional analyses, the
following changes were made relative to regional
results reported by these groups: for the IIASA
analyses, Oceanian OECD, China, India, and the
rest of Southeast Asia were added to create an
Asia-Pacific group; for the IEA analyses, Ukraine
was moved from Europe to Russia and the Former
95
Soviet Union and Mexico was moved from North
America to Latin and Central America. The
data from Harmsen et al. and the US EPA was
provided at relatively high resolution, allowing
its (re)assembly into the regions used in this
assessment. Total anthropogenic emissions are
estimated to be around 364–380 Mt/yr based on
the central values of bottom-up and top-down
estimates (Chapter 2). The IEA analysis does
not have total baseline emissions as it does not
include all sectors. The Harmsen analysis, using
the IMAGE model with all sectors, has a value
~390 Mt/yr, similar to the Chapter 2 estimates.
The IIASA analyses, without emissions from
forest fires and savannah burning or transport,
has baseline emissions of ~345 Mt/yr, similar
to the analogous value of ~355 Mt/yr based
on the source estimates described in Chapter
2. The US EPA baseline emissions, with the
same sectors as IIASA, other than the very
small source from agricultural waste burning,
are substantially lower at ~325 Mt/yr. This value
is based on national reporting to the UNFCCC,
supplemented with IPCC Tier 1 methods, and
this methodological difference is likely the main
driver of this much lower baseline. In order to
increase comparability across the mitigation
analyses and to bring the US EPA estimates in
line with the source estimates of Chapter 2, the
US EPA baseline and mitigation has been scaled
up by 7 per cent uniformly across all sectors,
which brings the total to the same amount as
the IIASA analysis, the next lowest baseline.
Given the large uncertainties in emission source
strengths discussed in Chapter 2, it is difficult
to classify a particular baseline as inconsistent
with observations however, and so this should
be regarded as primarily a method to enhance
comparability across the analyses. Based on the
findings of potential underestimates of fossil-
related methane sources based on ice cores
discussed in Chapter 2 (Hmiel et al. 2020), it
is noted that even the scaled US EPA baseline
estimates and the IIASA baseline estimates may
be biased low, with similar biases carrying over
to mitigation.
4.2.1 BROAD SECTORAL ANALYSES
To analyze the abatement potential across broad
sectors, these are grouped into five categories
for ease of visualization: all oil and gas, coal, all
livestock, rice, and all waste. Groups beginning
with all include multiple subsectors, whereas rice
is self standing and coal is related to mining alone,
though some analyses do divide this by active
and abandoned coal mines. These groupings
Global Methane Assessment / METHANE EMISSIONS MITIGATION
are somewhat arbitrary and were chosen in part
to reflect data availability – the IEA, for example,
does not include coal, precluding their inclusion
in a fossil fuel sector. Analysis of results within
these groupings is presented in Section 4.2.2.
At the global scale and considering abatement
measures regardless of costs, the four analyses
all find the greatest potential within the oil and
gas subsector, for which 2030 (2020 for IEA)
the maximum abatement range is 29–57 Mt/yr.
There is greater divergence in the cost estimates
than in the mitigation potential, with the average
abatement cost being around US$ 1 000 per
tonne of methane in the Harmsen analyses,
near zero in the IEA analysis, and having a net
negative cost of roughly US$ 700 per tonne
methane in the IIASA analysis. The average cost
in the US EPA analysis is nearly US$ 30 000
per tonne of methane, but this very high value
results from a small number of very expensive
measures. Restricting the analysis to those
measures with costs of less than US$ 25 000
(2018 US dollars) per tonne of methane, which
would be US$ 1 000 per tonne carbon dioxide
equivalent if using a global warming potential
over 100 years (GWP100) of 25, as in the US
EPA analysis, reveals 85 per cent of the total
abatement potential is available at an average
cost of US$2 300 per tonne of methane (Figure
4.5). Though this cost is much lower, it is still
higher than those of the other three analyses,
which do not incorporate abatement measures
costing more than US$ 25 000 per tonne of
methane.
The three analyses that included the waste
sector find very similar mitigation potentials of
29–36 Mt/yr but very different cost estimates
that range from the most expensive (Harmsen)
to the largest net negative costs (IIASA), for
which there are several reasons. Firstly, IIASA’s
mitigation potential considers almost full source
separation of organic waste with recycling or
energy recovery possible globally. US EPA
assumes a gradual shift away from open dumps
in developing countries but this appears to
be at a slower rate than in the IIASA analysis.
Secondly, in the cost estimates, IIASA considers
the cost-savings from recycling, energy recovery
and use, and the opportunity cost of avoided
landfilling. The cost-savings are subtracted
from the investment and maintenance costs to
arrive at net costs. Average world market prices
for recycled materials are used to value cost
savings of recycling and expected future gas and
electricity prices from IEA are used to reflect the
benefit of energy recovery in the waste sector.
The US EPA uses generally similar methods, but
96
the larger discount rate in their analysis would
make net costs higher. Note that the range
quoted above is for all measures regardless of
costs, but, as with the oil and gas subsector,
the US EPA analysis includes a small number
of very expensive measures in the wastewater
subsector. When these are excluded, the
mitigation potential in that analysis decreases
from 36 to 32 Mt/yr and the costs become
similar to the other analyses for the waste sector
(Figure 4.5).
As a result of the above, extremely large
discrepancies in the cost estimates result from
the use of differing boundaries within the sector,
different assumptions about global capacities,
potentially different timescales for phasing
in mitigation measures, and the inclusion of a
small set of extremely expensive controls. It
should also be noted that in the case of paper
recycling, in the IIASA estimate this is reflected
as a gain based the average global market
price of recycled paper, but this may inflate the
benefits because the market price of recycled
paper is often distorted by the pulp and paper
industry intervening in the market to keep the
price of both recycled and virgin pulp high and
stable, at least in some regions. The US EPA
analysis also includes a price for scrap paper of
US$ 140 per tonne.
In contrast, estimates of mitigation options for
livestock are in fairly good agreement as to the
costs but differ markedly in magnitude. In the
Harmsen analysis, the mitigation potential is 42
Mt/yr, making it the second largest potential,
the IIASA analysis is only 4 Mt/yr making it the
group with the smallest potential among the
five considered here, the US EPA analysis is in
between at 12 Mt/yr (Figure 4.5). Average costs
are similar at ~US$ 600 (IIASA) or ~US$ 1 000
(Harmsen and US EPA) per tonne of methane.
Results for abatement from coal mining are
relatively similar across the analyses in both
potential and costs, with values of 12–25 Mt/
yr1 and average costs of ~US$ 100–700 per
tonne. Finally, the mitigation potential is similar
in the three analyses for rice cultivation at 6–9
Mt/yr but, as with waste, costs diverge greatly
with the average value being ~US$ 150 per
tonne in the IIASA analysis and about 20 times
that, ~US$ 3 000 per tonne of methane, in the
Harmsen analysis and the US EPA in between at
US$ 1 700 per tonne.
The analysis of abatement potential available
at all costs is a useful way of characterizing the
maximum potential through targeted measures.
These results can be highly influenced by a
Global Methane Assessment / METHANE EMISSIONS MITIGATION
small subset of very expensive measures,
however, as noted previously for the US EPA’s
oil and gas subsector and wastewater results,
and as such it is also useful to examine those
measures that can be put into place at a low
cost, which is defined here as less than US$ 600
per tonne methane, which would correspond to
~US$ 21 per tonne of carbon dioxide equivalent
if converted using the IPCC Fifth Assessment
Report’s GWP100 value of 28 that excludes
carbon-cycle feedbacks.
The abatement potential of low-cost controls is
naturally smaller than the all-cost potential, but
the difference varies by source group and across
the four analyses. For oil and gas subsector, the
low-cost measures make up ~80 per cent of the
all-cost measures in the IIASA analyses, but only
~60 per cent in the US EPA and IEA analysis and
just 36% in the Harmsen analysis. This leads
to a large range of mitigation potentials, as in
the all-cost case, with low-cost abatement in oil
and gas ranging from 18 to 32 Mt/yr. Average
costs remain positive in the Harmsen analysis
despite the imposition of a low-cost threshold
that removes the most abatement, which
presumably largely accounts for its much smaller
fraction of the total abatement potential falling
within the low-cost threshold for the average
costs. Average costs are slightly negative in
the US EPA and IEA analyses, with IEA having
the largest mitigation potential, and strongly
negative for the IIASA analysis (Figure 4.6).
Around 10–20 Mt/yr in the waste sector can be
abated at low cost, though, as with all costs, the
differences in scope lead to a large divergence
in average costs across the analyses. They are,
however, consistent in finding that most coal-
related methane emissions abatement comes at
low cost, though the values range from 55 to 98
per cent of the total potential. In contrast, the
divergence between the three analyses expands
for the mitigation potential for livestock when a
low-cost threshold is imposed. Computing the
mean across the available analyses for each
subsector, the total mitigation potential available
at low cost is estimated to be about 73 Mt/yr.
This is a large fraction of the ~120 Mt/yr available
at all costs.
Including only those abatement measures
with negative costs reveals that the greatest
potential for net economic benefits without
accounting for environmental impacts comes
from mitigating emissions from the oil and gas
subsector and the waste sector (Figure 4.7).
This is unsurprising given that many of the
control measures for these two sources consist
of capture and use of natural gas. Potentials are
97
smaller in the coal, rice and livestock subsectors
and as noted previously, divergences in average
cost estimates are large for livestock.
Analysis of the nine regions used in this
assessment shows marked differences in
mitigation potentials. For the all-cost analysis,
the largest abatement potentials are as follows:
for Europe and India in the waste sector;
for China in the coal subsector, followed by
livestock; for Africa livestock, followed by oil and
gas; for Asia Pacific, excluding China and India,
the coal subsector and the waste sector, for the
Middle East, North America and Russia/Former
Soviet Union (FSU) the oil and gas subsector;
and for Latin America the livestock subsector.
As with the global analysis, a large majority of
these major abatement potentials can generally
be achieved at low cost. This is especially the
case for waste and the coal subsector, whereas
for livestock the costs are quite dependent
on the abatement analysis. The bulk of the
abatement in the oil and gas subsector can also
be achieved at low cost for North America in all
four analyses, but for Russia and the Former
Soviet Union this is the case in three of the four
with one finding only about a third of all cost
mitigation is available at low cost. For the Middle
East, two of the analyses found 70–80 per cent
of mitigation is available at low cost whereas
the other two found less than half was available.
Hence there are varying degrees of consistency
across the analyses, but in most regions, it is
clear where the largest methane abatement
potential lies.
The total mitigation potentials analyzed here are
a large fraction of the total emissions. For waste,
the ~30 Mt/yr all-cost reductions represent
roughly 40 per cent of the estimated 2030
emissions. For fossil fuels, the ~60 Mt/yr all-
cost mitigation potential is likewise about 40 per
cent of the estimated 2030 total. For agriculture,
however, targeted measures make a smaller and
less robust percentage contribution, with the
analyses providing estimates of about 10, 15
and 35 per cent of estimated 2030 emissions
from this sector (~10–50 M/yr).
This assessment focuses on the abatement
potential for methane in 2030. The mitigation
potential analyses of IIASA, Harmsen and the
US EPA extend further into the future, and we
also compare worldwide potentials in 2030
with those estimated for 2050. Though all three
analyses include projected changes in baseline
emissions and technological progress, there are
much larger differences in 2050 values relative to
their 2030 equivalents in the IIASA analysis than
Global Methane Assessment / METHANE EMISSIONS MITIGATION
in the Harmsen or US EPA analyses (Figure 4.8).
The greatest differences are seen in the waste
sector, in which, in particular for solid waste, the
IIASA analysis reports a more than three-fold
growth in low cost abatement potential between
2030 and 2050, whereas the growth is only 18
per cent in the Harmsen analysis that is based
on the 2013 US EPA estimates and 41 per cent in
the 2019 US EPA estimates. As noted previously,
this is primarily related to the assumption that
regions outside of Europe will continue to landfill
waste rather than use source separation. For
fossil fuels, there is again an 18 per cent growth
in the estimated low-cost abatement potential
over between 2030 and 2050 in the Harmsen
analysis and a similar 17 per cent growth in the
US EPA analysis whereas the growth is 67 per
cent in the IIASA analysis.
The IIASA analysis also includes more than
a tripling of the all-cost agricultural sector
abatement potential between 2030 and 2050,
which instead decreases slightly in the Harmsen
and US EPA analyses, bringing the results of
the IIASA and Harmsen analyses for this sector
closer together in 2050. The three analyses still
diverge substantially – 20 Mt in the US EPA
analysis, 35 Mt of methane according to IIASA
and 68 Mt according to Harmsen. They are
closer for rice cultivation with 8 Mt in the US EPA
analysis, 15 Mt in Harmsen, and 16 Mt in IIASA,
but there are large differences in the abatement
of livestock emissions; 11 Mt in the US EPA
and 16 Mt in IIASA while Harmsen suggests
53 Mt. The difference is in the assessment of
the feasible abatement potential for enteric
fermentation in ruminants. While Harmsen sees
98
a large potential from replacing low-productivity
indigenous cows and cattle globally with
imported highly productive breeds, the other
analyses only see this as a feasible solution for
large industrial farms. For smallholder farmers
in Africa and Southeast Asia, they do not see
this as a feasible targeted solution without first
implementing institutional and social reforms.
This is because of the important role that
keeping livestock plays in poor farmers’ risk
management; as assets when access to credits
and health insurance schemes is missing.
Changes in average abatement costs for
the low-cost measures are small over the
2030–2050 period in all three analyses for the
fossil fuel and waste sectors, as well as their
underlying subsectors. Projected changes in
low-cost abatement potentials are substantial
in the agricultural sector in the IIASA analysis
while they are small for the other two, but costs
in this sector are also not projected to change
greatly over time. The large divergence in costs
therefore remains with the average costs of low-
cost measures across all sectors having fairly
small positive values, from US$ 15–200 per
tonne of methane, in the Harmsen study which
had a minimum cost of zero whereas they are
negative in the IIASA and US EPA analyses in
which the range is about -US$ 1 000–10 000
per tonne for the global sectoral averages for
waste, fossil fuels and agriculture. In 2050, the
total low-cost abatement potential is estimated
to be about 110 Mt/yr for the average of the
three analyses, roughly 50 per cent larger than
the 2030 potential.
Global Methane Assessment / METHANE EMISSIONS MITIGATION 99

Figure 4.5 Worldwide methane mitigation potential by sector, million tonnes, versus average mitigation cost, 2018
US$ per tonne, from the indicated analyses

Note: All costs of mitigation are included below a US$ 25  000 per tonne of methane exclusion
threshold.

Source: UNEP and CCAC

Figure 4.6 Worldwide methane mitigation potential by sector, million tonnes, versus average mitigation cost, 2018
US$ per tonne, from the indicated analyses

Note: Only includes abatement measures that cost less than 2018 US$ 600 per tonne of methane.

Source: UNEP and CCAC

 Global Methane Assessment / METHANE EMISSIONS MITIGATION 100

Figure 4.7 Worldwide methane mitigation potential by sector, million tonnes, versus average mitigation cost, 2018
US$ per tonne, from the indicated analyses

Note: Only includes abatement measures with negative net costs.

Source: UNEP and CCAC

HARMSEN (2030) IIASA (2030) US EPA (2030)

80
HARMSEN (2050) IIASA (2050) US EPA (2050)
70

60
METHANE (Mt)

50

40

30
•

20

10

0
CK

RE

AS

TE

TE
EL

E
AS

AS
G
U

AT
O

FU
LT
ST

&

W
W
CU

IL

IL
VE

TE
D

L
O

AL
SS

LI

AS
RI
LI

SO
FO
AG

AL
L

W
AL

L
L

L
L

AL
AL

AL
AL

Figure 4.8 Comparison of mitigation potentials in the Harmsen, IIASA and US EPA analyses by sector for measures
costing less than 2018 US$ 600 per tonne methane, 2030 and 2050

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.2.2 DETAILED SUBSECTOR ANALYSES
Within the three of the five categories examined in
the previous section, all oil and gas, all livestock,
and all waste, there are several subsectors with
different abatement potential, techniques and
costs. These are explored here, beginning with
oil and gas, putting this into context with the
fossil sector as a whole, followed by waste and
then livestock, similarly putting this into context
with the agricultural sector as a whole.
The IEA analysis, which is only available
for the oil and gas subsector, estimates
that the full application of best available
abatement technologies could reduce the oil
and gas subsector’s methane emissions by
approximately 57 Mt annually, 72 per cent of
its current emissions. These results are similar
to those reported in the Harmsen study with
substantially lower potential according to the
IIASA and US EPA analyses (Figure 4.9). Within
the oil and gas subsector, the IEA analysis finds
larger potential in gas whereas the Harmsen and
IIASA analyses find larger potential in oil, in the
Harmsen case substantially so. The US EPA
analysis does not separate oil and gas.
In comparison, mitigation options in the coal
sector are slightly smaller than in either the oil
or gas subsectors in the IIASA and Harmsen
analyses, but the coal subsector potential is
larger in the US EPA analysis than in either of
those studies. The average of these leads to
roughly comparable contributions from oil, gas
and coal in the total abatement potential. The
total fossil fuel mitigation potential is fairly robust
across the three analyses, with the full range
captured as 53 ± 7 Mt/yr. The average of the
three coal subsector mitigation analyses and the
four oil and gas analyses is higher at 58 Mt/yr.
This represents from roughly 30 per cent to just
over half of projected 2030 emissions from this
sector. Looking at only the low-cost measures,
the total fossil fuel abatement potential differs
more sharply across the three analyses, with
values of 24–42 Mt/yr (Figure 4.10) and an
average based on the three coal subsector
mitigation analyses and the four oil and gas
analyses of 37 Mt/yr. Again, it seems that the
abatement potential for the low-cost subset of
all controls is similar for the three fossil fuels,
except in the Harmsen analysis which includes
little low-cost abatement within the oil subsector.
Specific controls for all sectors applied in the
various analyses are listed in Table 4.1.
The IEA, IIASA and US EPA analyses, but not
the Harmsen study, all analyzed the portion
101
of identified emissions reductions that could
be achieved at a net cost savings. These are
particularly large for the oil and gas subsector
in which there is a net saving to operators
through reduced losses. For this subsector, the
portion of total reductions (Figure 4.9) that can
be realized at negative cost is 53 per cent in the
IEA analysis, 55 per cent in the IIASA analysis,
and 25 per cent in the US EPA analysis for the
global total in 2030. Though the IEA and IIASA
analyses are similar for the global total, they
vary substantially for some regions. In Europe,
for example, the portion of abatement available
at negative cost is only 31 per cent in the IEA
analysis, but 87 per cent in IIASA’s analysis.
The spread is also large for North America, with
values of 56 per cent in the IEA analysis and 92
per cent in the IIASA analysis. Russia and the
Former Soviet Union has a generally smaller
percentage of abatement potential with negative
costs, 45–59 per cent, whereas all other regions
have more than 70 per cent of emissions abatable
at negative costs in at least one analysis. For
the coal subsector, IIASA and US EPA, the two
analyses with results for the subsector, find that
3 (US EPA) to 17 (IIASA) per cent of identified
global emissions reductions can be achieved
at negative costs, and again results are more
variable at the regional level. For the waste
sector, these two analyses differ considerably,
with 27 per cent of global emissions reductions
having negative costs according to the US EPA
analysis and 62 per cent in the IIASA analysis,
due to the differing scope of their economic
analysis previously discussed.
Additional detail is provided for the oil and
gas subsector within both the IEA and IIASA
analyses. As shown in Figure 4.9, the IIASA
analysis finds substantially less mitigation
potential in the oil and gas subsector than
the other three analyses, including the IEA.
Examination of the components of this subsector
shows that the differences are consistent across
the components, but the relative differences are
largest for natural gas production and smallest
for oil production with downstream gas in the
middle (Figure 4.11). Some of this discrepancy
comes from the inclusion of technology turnover
times by the IIASA that is not included in the
IEA analysis, as can be seen by examining the
IIASA abatement potential in 2050 which is
substantially larger as roughly two thirds of the
total mitigation potential in that analysis takes
place in the 2030s with only on third in the
2020s. The difference between the oil and gas
totals in the IIASA and US EPA results relative
to the Harmsen results may also be linked to
Global Methane Assessment / METHANE EMISSIONS MITIGATION
turnover times, as the Harmsen analyses are
based on assessments that are 5–10 years
older than the IIASA and US EPA assessments.
Harmsen is a review of assessments produced
around 2012–2016 while the US EPA and IIASA
studies were for 2019 to 2020. This leaves only
a ~10-year horizon until 2030 in those analyses
versus 15–18 years in the Harmsen study.
Individual abatement technologies were
examined at a very detailed level in the IEA
analysis. The IEA classifies all emissions from
production, gathering and processing of fuels
as upstream and all emissions from refining,
transmission and distribution as downstream.
Their analysis shows that the category with
the largest mitigation potential, upstream
leak detection and repair (LDAR), is also the
cheapest. The bulk of the negative cost benefits
are obtained from five abatement technologies,
each of which contributes substantial health
and climate benefits while simultaneously
leading to economic gains even when the value
of the environmental impacts is not included
(Figure 4.12). The majority of the negative
cost options occur in four source categories,
onshore conventional oil and gas and offshore
oil and gas, whereas abatement within the
unconventional oil and gas and downstream
categories typically have positive costs. Full
application of all abatement technology in the
offshore and onshore conventional categories,
including individual applications with positive
costs, would result in abatement of ~40 Mt/yr
of methane emissions, leading to a reduction in
global mean surface warming of ~0.05 C and
~45 000 avoided premature deaths annually
with an overall negative cost for each of these
four production sources – Chapter 5 includes
more discussion of socio-economic impacts of
abatement. Accounting for the environmental
benefits of emissions reductions, all the
identified IEA targeted measures have negative
net costs. It should be note, however, that the
most costly of the measures analyzed by IEA,
the installation of flares, would likely lead to
increased emissions of black carbon that would
have damaging health and climate impacts
(Stohl et al. 2015).
Within the waste sector, all cost abatement
potential is concentrated within the solid waste
subsector which has three to six times the
potential found in the wastewater (sewage)
subsector (Figure 4.9). Totals in the three available
analyses are very similar for the full waste sector,
so that the full range is captured by 32 ± 4 Mt/
yr. Hence this sector has about half the potential
of the fossil sector for all cost measures and a
102
much narrower uncertainty range. Evaluating
this mitigation potential as a share of projected
2030 waste sector emissions is complicated by
a large divergence between them, which were
~70 Mt/yr in the Harmsen and US EPA analyses,
whereas there was a much larger value of 114
Mt/yr in the IIASA analysis. Hence although all
the studies find similar abatement potential, the
share of 2030 emissions from waste estimated
to be abatable ranges from just 25 per cent in
the IIASA analysis to ~40-50 per cent in the
US EPA and Harmsen analyses. For low-cost
measures in the waste sector, the analyses are
again fairly consistent with all falling within the
range 16 ±5 Mt/yr.
Within the waste sector, the IIASA analysis
provides additional detail on specific measures.
The largest abatement within the analysis is
for municipal waste, for which the separation
and treatment of biodegradable municipal
waste with no biodegradable waste being
sent to landfill forms the primary component
of the estimated potential abatement. The
separation of biodegradable waste is another
measure with a negative cost, even without
accounting for environmental impacts, with an
estimated average value per tonne of methane
of ~US$ 8 500 per tonne methane and a global
mitigation potential of ~16 Mt/yr. Industrial
solid waste also shows significant potential for
abatement, ~5 Mt/yr, with again a large majority
of that able to be achieved at zero or net negative
costs – the average cost is -US$ 6 800 per tonne
of methane. It should be noted that, in the IIASA
analysis, the vast majority of the waste sector’s
methane emissions, other than the relatively
small subsector of domestic wastewater, can
be eliminated by targeted control measures by
2040. The US EPA analysis categorizes waste
sector measures differently, but similarly finds
many with negative cost. The largest mitigation
potential with a negative net cost is nearly 9
Mt/yr for gas capture and usage for electricity
generation at an average cost -US$ 1 750 per
tonne methane. The largest of the negative
costs is for landfill gas recovery for direct use,
with an average value of -US$ 3 100 per tonne
and an estimated mitigation potential of 2.2
Mt/yr. Among other waste sector measures in
the US EPA analysis, flaring of landfill gas and
composting are on average low-cost measures,
whereas anaerobic digestion, waste to energy,
mechanical biological treatment and paper
recycling all have average costs higher than this
assessment’s low cost threshold.
Three of the analyses included the agriculture
sector, and as noted previously they differ greatly
Global Methane Assessment / METHANE EMISSIONS MITIGATION
from one another. These large differences stem
from livestock, as the analyses are in fairly close
agreement with mitigation potentials for rice of 6,
9 and 9 Mt/yr for the all-cost abatement potential
(Figure 4.9), corresponding to ~20–30 per cent
of emissions from this subsector. The analyses
differ markedly on the costs associated with rice
methane mitigation measures, however, with
nearly all the identified potential achievable at
low cost in the IIASA analysis, 55 per cent in the
US EPA analysis, and less than a third of them in
the Harmsen analysis (Figure 4.10). For livestock,
the all-cost emissions reductions represent 3
per cent of that subsector’s projected 2030
emissions in the IIASA analysis, 8 per cent in the
US EPA analysis and 30 per cent in Harmsen’s
(Figure 4.9), with the differences arising primarily
from differing assumptions about the previously
discussed plausibility of introducing high
productivity breeds to Africa and Southeast Asia
as for both rice and livestock the three analyses
project similar 2030 baseline emissions. For the
small agricultural waste-burning subsector within
agriculture, targeted measures are estimated to
be able to eliminate emissions completely at no
cost in the IIASA analysis, but they will require
regulation and enforcement.
Within the livestock subsector, the analyses
include changes in management practices and
in animal feed. Changes in management may
include improving herd health, breeding for
improved productivity, and shifts from pastoral
to intensive systems for cattle. In the last,
additional impacts on biodiversity, those living
near intensive operations, and the treatment of
the animals themselves need to be considered
and are likely to be highly regionally dependent
(Gerber et al. 2013). In the case of animal feed,
options include processing feed grain to improve
digestibility and feed supplements including
nitrate and tannins (as antimethanogens).
Additional feed supplements offer potential
for reducing methane emissions from the
livestock sector but are currently considered
experimental and are therefore not included
in most abatement potential analyses. In one
study, an artificial supplement given to dairy
cows reduces methane production from enteric
fermentation by 30 per cent with no adverse
effects on milk production and a gain in overall
cattle growth (Hristov et al., 2015). Widespread
application of this practice could lead to
significant abatement of methane, potentially of
up to 20 Mt/yr (Shindell et al. 2017), though the
supplement has thus far been tested in only a
single small herd. Adding small quantities of the
seaweed Asparagopsis taxiformis, potentially in
103
combination with Oedogonium sp., to ruminant
feed has also been found to greatly reduce
methane production from enteric fermentation,
with decreases of 80 per cent in sheep and up
to 99 per cent in cows based on in vitro trials
(Roque et al. 2021; Abbott et al. 2020; Vijn et al.
2020; Kinley et al., 2016; Li et al., 2016; Machado
et al., 2016; 2018). In both cases, further studies
are needed to confirm benefits and explore
the potential for scaling up these practices as
well as fully characterizing potential effects on
animal health and the long-term persistence of
the identified potential benefits.
All three mitigation analyses for the livestock
sector include methane emissions abatement
through improved manure management. The
primary policy in this area is the adoption of
farm-scale anaerobic digesters for manure from
cattle and pigs. There have also been analyses
of the potential of additives mixed into the
manure slurry ponds to reduce emissions, with
both widely available acids such as sulfuric or
lactic acid (Sokolov et al. 2019; Sommer et al.
2017; Petersen et al. 2014) and commercial
products (Peterson et al. 2020; Borgonovo et al.
2019) having been explored. Both seem to offer
the potential to greatly reduce both methane
and ammonia emissions based on field studies
and sulfuric acidification has been widely used
in Denmark primarily to control ammonia. In
the rest of the world, however, there has been
limited use of such techniques, primarily owing
to concerns over the safe handling of the acids
and uncertainty regarding the long-term impacts
on soils. In terms of mitigation potential, it
appears that such practices are likely to
achieve roughly the same methane reductions
as the application of well-managed anaerobic
digestion with biogas recovery that is included
in current mitigation analyses. These are
therefore not included as separate measures in
the IIASA analysis, although they are included
in the Harmsen analysis, along with anaerobic
digesters, decreased manure storage time,
improved manure storage covering and improved
housing systems and bedding (Table 4.1).
It is noted that many emissions abatement
measures are expected to affect methane while
leaving other emissions virtually unchanged;
such measures include capture of fugitive
emissions or controls on landfills, use of
anaerobic digestion for manure, and introducing
intermittent irrigation practices for rice. Other
measures, however, in particular those that affect
people’s diets or, especially, livestock are likely
to affect emissions of carbon dioxide and nitrous
oxide, both greenhouse gases – for example,
 Global Methane Assessment / METHANE EMISSIONS MITIGATION 104

the slurry acidification of manure discussed
previously may increase nitrous oxide emissions.
The impacts of those additional greenhouse gas
emissions on global climate could be accounted
for fairly easily based on prior modelling of the
response to well-mixed greenhouse gases.
Changes such as fuel switching or efficiency
increases would also affect multiple pollutant
emissions, and hence are typically studies with
IAMs rather than using MAC curve analyses –
efficiency is discussed further in Section 4.3.1.
In addition to methane, fuel switching from coal
or increasing energy efficiency would reduce
carbon dioxide and volatile organic compound
emissions and air pollution, switching away from
gas would reduce carbon dioxide and volatile
organic compound emissions (Fann et al. 2018),
and reducing consumption of cattle-based
foods would be likely to reduce both carbon
dioxide emissions by reducing deforestation,
and nitrous oxide emissions through reducing
fertilizer use, and hence all these measures
would be expected to yield addition benefits
through non-methane emissions changes rather
than having additional impacts that offset some
of those related to planetary warming described
in Chapter 3.

180

160

140

120
METHANE (Mt)

100

80
•

60

40

20

0
CE

CK

RE

AL

IL

AS

IL

TE

E
EL

T
E
O

AS

AS
CO

G
U

AT
RI

FU
&
LT
ST

W
W
AS
CU

IL
VE

TE
D

L
AL
G

SS

LI
RI

AS
LI

SO
FO
AG

RA
L

W
AL

L
U
L

AL
AL

AT
N

BASELINE 2030 HARMSEN (2030) US-EPA (2030) IIASA (2030) IEA (2020)

Figure 4.9 Methane emissions mitigation potentials for 2020 for the IEA analysis and 2030 in the Harmsen,
IIASA and US EPA analyses including all cost measures as well as the average baseline 2030 emission projected in
the IIASA, Harmsen and US EPA analyses

Note: The potential in the IIASA and US EPA studies is based on a 10 year time horizon whereas the
Harmsen study has an ~15 year horizon.

Source: UNEP and CCAC

 Global Methane Assessment / METHANE EMISSIONS MITIGATION 105

50

40
METHANE (Mt)

30

Figure 4.10 Methane

emissions mitigation
•

20
potentials for 2020 for
the IEA analysis and
2030 in the Harmsen,
IIASA and US EPA
analyses for measures
10 costing less than 2018
US$ 600 per tonne of
methane

Source: UNEP and

CCAC
0
CE

CK

RE

AL

IL

AS

AS

TE

TE
EL

E
O

AS

AS
CO

G
U

AT
RI

FU
LT
ST

W
W
AN
CU

IL
VE

TE
D

L
AL
SS

LI
RI

AS
LI

IL

SO
FO
AG

O
L

W
AL

L
AL
L

AL
AL

US-EPA (2030) HARMSEN (2030) IIASA (2030) IEA

OIL
PRODUCTION

NATURAL GAS
PRODUCTION

DOWNSTREAM
GAS IIASA (2030) IEA (2020)

0 5 10 15 20 25 30

• METHANE (Mt)
Mt METHANE

Figure 4.11 IEA estimated mitigation potentials in the oil and gas sectors, million tonnes of methane

Note: IEA potentials are estimated for 2020 while the IIASA values are shown for 2030.

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION 106

Figure 4.12 Relative savings/costs (2018 US$ per tonne) of targeted methane abatement measures in the oil and gas
sector identified by the IEA showing the reduction potential, ‘000 tonnes; impacts on premature deaths and global
mean surface temperature, degrees Centigrade

Source: UNEP and CCAC

Global Methane Assessment / METHANE EMISSIONS MITIGATION
Table 4.1 Emissions control measures included in at least one of the mitigation analyses
FOSSIL FUELS
Oil and gas: upstream and
downstream leak detection and
repair (LDAR).
Oil and gas: blowdown capture;
recovery and utilization of vented
gas with vapour recovery units and
well plungers; installation of flares.
Oil and gas by existing devices:
replace pressurized gas pumps
and controllers with electric or air
systems; replace gas-powered
pneumatic devices and gasoline or
diesel engines with electric motors;
early replacement of devices with
lower-release versions; replace
compressor seals or rods; cap
unused wells.
Coal mining: pre-mining
degasification; air methane
oxidation with improved ventilation.
Coal mining: flooding abandoned
mines.
BEHAVIOURAL AND TECHNOLOGICAL CHANGES
FOSSIL FUELS
Fuel switching from fossil fuels to
renewables/nuclear.
Energy demand management.
Energy efficiency improvement.
Emissions pricing.
TECHNICAL CONTROLS
WASTE
Municipal solid waste: source
separation with recycling/reuse; no
landfill of organic waste; treatment
with energy recovery or collection
and flaring of landfill gas.
Industrial solid waste: recycling or
treatment with energy recovery; no
landfill of organic waste.
Residential wastewater: upgrade
of primary treatment to secondary/
tertiary anaerobic treatment with
biogas recovery and utilization.
Wastewater treatment plants
instead of latrines and disposal.
Industrial wastewater: upgrade of
treatment to two-stage treatment,
Agricultural waste burning: ban and
i.e., anaerobic treatment with
enforcement of existing bans.
biogas recovery followed by aerobic
treatment.
WASTE
Emissions pricing.
107
AGRICULTURE
Cattle, sheep and other ruminants
through enteric fermentation:
feed changes and supplements;
breeding to improve productivity
and animal health/fertility.
Ruminants and pigs through
manure management: treatment
in biogas digesters; decreased
manure storage time; improved
manure storage covering; improved
housing systems and bedding;
manure acidification.
Rice cultivation: improved water
management or alternate flooding/
drainage wetland rice; direct
wet seeding; phosphogypsum
and sulphate addition to inhibit
methanogenesis; composting rice
straw; use of alternative hybrids.
AGRICULTURE
Reduced food waste.
Dietary change.
Emissions pricing.
Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.3 ABATEMENT IN RESPONSE TO
METHANE-SPECIFIC POLICIES
4.3.1 MULTIPLE INTEGRATED ASSESSMENT
MODEL 2050 ANALYSES
The IAMs used to generate scenarios consistent
with particular warming targets are designed to
achieve those targets at the lowest possible cost
considering all sectors, usually all pollutants, and
all mitigation options including targeted controls,
behavioural changes and directed policies such
as taxes. They therefore include many methane
abatement measures, as discussed in Section
4.1. Here the methane abatement in those
scenarios are compared with the total potential
from the analyses discussed in Section 4.2 and
with the response seen in IAMs to methane-
specific policies.
An analysis of nine IAMs has probed their changes
in methane emissions in response to various
policies, focusing on the year 2050 (Harmsen et
al. 2019b), and here the impact of low carbon
scenarios on methane in 2050 is also discussed.
In the reference cases used in the Harmsen et al.
(2019b) study, which follow SSP2, anthropogenic
emissions are projected to rise from 2010 levels
of about 360 Mt/yr to about 460 Mt/yr in 2050.
Transitioning to the least-cost mitigation pathway
consistent with the 2ºC climate target in those
models leads to a decrease in emissions to about
185 Mt/yr in 2050 to, a reduction of about 275
Mt/yr, 60 per cent. This is towards the high end
of the emissions consistent with 1.5ºC scenarios
shown in Section 4.1, which range from about
60 to 200 Mt/yr in 2050. In contrast, were a
climate policy to consider mitigation of carbon
dioxide alone, i.e. following a 2ºC-like scenario
but imposing an emissions price only on carbon
dioxide, the reduction would be only about 80 Mt/
yr. Comparison of these results demonstrates
that the bulk of the methane reductions under a
2ºC scenario come about from policies targeted
directly at methane as opposed to being merely
a co-benefit of a strategy focused of carbon
dioxide. This highlights that general measures
to decarbonize the economy result in only a
modest reduction in methane emissions over
the near term. In the case of SSP2, the carbon
dioxide focused mitigation policies lead to only
~30 per cent of the methane abatement in 2050
seen under a broad 2ºC scenario, and a similar
fraction over the longer term with values of 22–48
per cent of the 2ºC abatement in 2100 (Harmsen
et al. 2019b). This stems from the limited impact
of carbon dioxide-focussed mitigation policies on
waste and agricultural sector methane emissions
as a whole and on end-of-pipe controls for
108
methane on remaining fossil fuel use within the
energy sector.
Across sectors, under a 2ºC climate policy
the models produce reductions in methane
emissions from fossil fuels that are essentially
as large as their maximum reduction potential.
In contrast, reductions in emissions from
agricultural subsectors – manure, enteric
fermentation and rice cultivation – are lower than
the maximum potential, whereas those for waste
– sewage and landfills – are at the maximum in
some models but lower in others. Reductions are
typically greater than 50 per cent for fossil fuels
and landfills, and they extend as high as 95 per
cent, but smaller for other sectors, with 25–55
per cent for manure, 15–80 per cent for sewage,
25–65 per cent for rice and about 15–45 per
cent for enteric fermentation. Along with these
percentage changes, it is important to consider
the overall magnitude of each source. Along a
2ºC pathway in the IAMs, emissions from manure,
oil, coal and landfills are nearly always 20 Mt/yr or
less each by 2050, and emissions from sewage,
rice and natural gas are nearly always 35 Mt/yr or
less respectively, whereas emissions from enteric
fermentation primarily range from about 80 to 140
Mt/yr. Hence enteric fermentation, especially from
cattle, becomes by far the dominant remaining
source of methane emissions in the IAMs under
2ºC scenarios, even in those with the largest
estimates of reduction in that sector – emissions
in the very lowest emission example for this sub-
sector are slightly more than 60 Mt/yr.
The analysis of methane emissions in IAMs not
only explored the response in least-cost multi-
pollutant pathways and in the carbon dioxide tax
only pathways, but also examined the role of a
cross-sector focus on methane (Harmsen et al.
2019b). This was accomplished by imposing
an economy-wide tax on methane emissions
and examining how the models responded.
In principle, this is similar to imposing a cap
on methane and allowing markets to find the
least-cost way of staying within that cap. The
tax was imposed from 2020 at a value of 2018
US$ 2 100 × 1.05 to the power of (year-2040)
per tonne of methane. This leads to ~US$ 790
in 2020 and ~US$ 1 300 in 2030 greater than
the low-cost threshold. Hence these analyses
encompass reductions larger than those included
in these analyses of marginal abatement potential
available at low cost and that covering all costs
reported in Section 4.2.
The results show that the models generally respond
to a methane tax by most strongly reducing
emissions from the energy sector, followed by
 Global Methane Assessment / METHANE EMISSIONS MITIGATION 109

land use, largely agriculture, and lastly the waste
sector (Figure 4.13, top). Differences between the
models are very large, however, especially for
the land-use and waste sectors. Delving more
deeply into the IAM responses to a methane
tax shows that the models typically produce the
largest total abatement from the coal subsector,
followed, in order, by enteric fermentation,
landfills, gas, sewage, rice and oil (Figure 4.13,
bottom). Contributions from reductions in energy
demand, manure management and agricultural
waste burning are small in nearly all models.
Model-to-model differences are considerable,
however, especially for the fossil fuel subsectors,
enteric fermentation and sewage. The AIM model
is particularly distinct, with far larger mitigation,
METHANE (Mt)
~20–30 Mt/yr in the enteric fermentation, sewage,
oil and agricultural waste burning subsectors
than any other IAM.
Across sectors, the models on average reduce
2050 emissions by a total of 209 Mt/yr in
response to the imposed tax, representing a
reduction of about 50–75 per cent in the projected
2050 emissions. The upper end of this range is
consistent with the methane reductions in 2ºC
pathways in these same models, though on
average the tax produces a smaller reduction than
that which occurs in the 2ºC pathways. Naturally
these results would be sensitive to the level of the
tax imposed in the models, though it is expected
that the relative sectoral contributions are fairly
robust to the level of the tax.

Figure 4.13 Abatement

in 2050, relative to the
baseline, by sector in
integrated assessment
model simulations that
imposed a methane
tax, million tonnes of
methane

ALL ENERGY ALL LAND USE ALL WASTE Note: The horizontal
lines within each
box are the median
values, open circles
are average values,
filled symbols are
individual integrated
METHANE (Mt)

assessment model
results, and the box
and uncertainty bar
show the 25–75th
percentile and full
•

range, respectively.

Source: UNEP and

CCAC
D

S
AL

IL

AS

RE

LL
CE

E
AN

IN
O

AG
O

FI
CO

U
RI

TI
EM

RN
AN

W
TA

SE
D

BU
M
EN

LA
Y
G

TE
RM
ER

AS
FE
EN

W
C

L
RI

RA
TE

U
EN

LT
CU
RI
AG
Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.3.2 2030 ANALYSIS USING THE GLOBAL
CHANGE ASSESSMENT MODEL
Though the use of multiple IAMs in Harmsen
et al. (2019b) provides valuable information
about the relative ease of reducing methane
emissions from different sectors, that analysis is
restricted to 2050 and 2100 and only explored
SSP2. Here, additional analyses were therefore
performed using an updated version of one
of the IAMs included in that study, the Global
Change Assessment Model (GCAM), to examine
the response in 2030 across multiple SSPs
for consistency with the mitigation potential
analyses. The GCAM is an IAM that is capable
of simulating the dynamics of human-Earth
systems and how those systems respond to
global changes, such as in population, economic
output, energy resource availability, technology
performance and climate. The version used
was GCAM5.2, as opposed to the version in
Harmsen et al., GCAM4, that was also used to
generate SSPs. The GCAM5.2 is an updated
version of GCAM5.1, described in Calvin et
al. (2019). While earlier versions of the GCAM
model focussed more heavily on the energy
sector, GCAM 5.2 contains additional detail in
other sectors and can better reflect the reality
of human-Earth systems. Compared to GCAM4,
GCAM5.2 includes a new climate model, new
land region and new representations of water
supply and demand, regional agricultural
markets, and depletable energy resources.
In this study, the environmental and socioeconomic
impacts of a global tax on methane are explored.
Two baseline scenarios that assume no specific
climate policy is in effect were simulated, as well
as versions of those scenarios in which the same
global methane tax as in Harmsen et al. (2019b) is
imposed. The differences between baseline and
tax scenarios show the effects of the imposed
tax for each baseline. The first baseline scenario
follows the second shared socioeconomics
pathway (SSP2), as in the IAM simulations in
Harmsen et al. (2019b). No major changes from
historical patterns in social, economic and
technological trends are assumed in SSP2. To
examine the sensitivity to the choice of baseline
scenario, the second baseline follows SSP1, a
sustainability scenario.
Differing from many other climate policy
scenarios that either impose a tax on carbon
dioxide or carbon dioxide equivalent, the
methane policy scenario involves a global tax
specifically on methane that increases by 5 per
cent annually. The GCAM separates methane
emissions into three categories:
110
1. methane emissions from agricultural
production, including but not limited to rice
cultivation and livestock;
2. methane emissions from agricultural waste,
including the burning of crop residues; and
3. a general category which covers all other
methane emissions.
The methane tax is applied to all three categories
simultaneously. The GCAM includes marginal
abatement curves for many methane sources,
allowing the methane emission factor to be
reduced as a function of the magnitude of the
methane tax. It also has an option that allows
the effectiveness of controls to improve time,
representing technological improvement. This
analysis opted not to use this option as the
resulting removal efficiencies for some control
options approached 100 per cent for future years,
exceeding what was believed to be reasonable.
As a result, it is expected that the scenarios used
here may have underestimated methane mitigation
potential and overestimated costs. To analyze the
results, methane emissions were aggregated into
seven groups: agriculture, fossil-fuel extraction,
buildings, electricity, industry, transportation and
waste, allowing the exploration of the tax’s effect
on different sectors.
It was found that implementing the methane tax
described above (2018 US$ 2 100 rising by 5 per
cent a year up to 2040 per tonne of methane)
in GCAM to represent a policy specifically
designed to reduce methane emissions causes
an immediate drop in anthropogenic methane
emissions in the implementation year, 2025,
and, under the conditions of SSP2, a fairly
constant emission rate thereafter (Figure 4.14).
As emissions continue to grow in the reference
case, the influence of the tax grows with time,
so that anthropogenic emissions are reduced by
20 per cent in 2030, 25 per cent in 2050 and 40
per cent in 2100.
The abatement in response to the economy-wide
methane tax under SSP2 comes primarily from
the fossil-fuel sector in the GCAM (Figure 4.15).
Within this sector, abatement through 2030
is largely achieved by reductions in emissions
related to coal which provide 43 Mt/yr of
abatement relative to 15 Mt/yr for gas and 6 Mt/yr
for oil. Reductions in emissions in the agricultural
and waste sectors are also substantial, though
smaller. Within agriculture, these come primarily
from abatement of emissions related to rice
cultivation, 10 Mt/yr of abatement, with an
additional 7 Mt/yr from cattle and 4 Mt/yr each
from smaller ruminants –pigs, sheep and goats.
Global Methane Assessment / METHANE EMISSIONS MITIGATION
The waste sector has a total reduction of 9 Mt/
yr entirely from municipal landfills. An additional
minor contribution of less than 1 Mt/yr globally
in 2030, comes from reductions in the building
sector. The total drop in 2030 in response to
the tax is 92 Mt/yr. Examining the regional
breakdown of responses to the tax, the largest
share of the reductions takes place in China,
24 per cent; and Russia and the Former Soviet
Union, 24 per cent; followed by Europe, 20 per
cent. In contrast, 4 per cent or less each of the
reductions come from North America, South and
Central America, India and the Middle East, with
Africa and Asia Pacific in between at 11 per cent
and 12 per cent, respectively.
In contrast, under SSP1, emissions of methane
are markedly lower in the baseline scenario and
the imposition of the methane-specific tax has
less impact on the emissions, decreasing them
by 70 Mt/yr in 2030. Under SSP1, the majority
of the emissions reductions again come from
the fossil-fuel sector, which provides 58 Mt/yr
of the 70 Mt/yr reduction. This is 91 per cent
of the reduction under SSP2 but represents 83
per cent of the total reductions, a larger fraction
than the 70 per cent contribution from the fossil
sector under SSP2. The larger fraction of the
total coming from the fossil-fuel sector under
SSP1 largely stems from the greater baseline
reductions in the agricultural sector under that
scenario. Those remove nearly all the methane
abatement potential in that sector that responded
to the methane tax under SSP2, i.e. the rice
and livestock abatement, so that under SSP1
the agricultural sector emissions are largely
unchanged by the imposition of the methane tax
(Figure 4.15). Note that the methane reduction
under the baseline SSP1 scenario in the GCAM
is slightly greater than that in the SSP1 marker
scenario (Figure 4.1), but only marginally so.
Methane emissions reductions in the waste
sector in response to the tax are 9 Mt/yr- in 2030
under SSP1, equal to those under SSP2.
For comparison with the analysis from Harmsen
et al. (2019b), presented in Section 4.3.1,
the sectoral contributions in 2050 were also
examined. Under SSP2, GCAM5.2 produces
methane emissions abatements of 81 Mt/yr in
the fossil fuel sector, 32 Mt/yr- in the agricultural
sector and 15 Mt/yr in the waste sector. In
response to the same tax and using the same
SSP2 baseline, GCAM4 produced abatement
values of 96, 29 and 18 Mt/yr in the fossil-fuel,
agriculture and waste sectors, respectively.
Hence overall the updated version of GCAM
behaves fairly similarly to the prior version,
though GCAM4 appears to have included more
111
use of coal in its baseline scenario. As shown
in Figure 4.13, the abatement within the fossil
fuel sector in response to a methane tax in the
GCAM is right in the middle of that obtained
across the many IAMs that performed these
simulations. In 2050, the abatement within the
fossil-fuel subsectors is 58 Mt/yr from coal, 16
Mt/yr from gas and 6 Mt/yr from oil, the latter
two being almost the same as the 2030 values.
In comparison with other IAMs, these results
put GCAM5.2 towards the high-end for coal
abatement but at the low end for the other fuels.
The abatement in the agricultural. Land-use and
waste sectors, however, is the lowest of all the
participating models with a value roughly half
the multi-model mean and less than a quarter
of the highest value. Still examining 2050,
GCAM5.2 produces an abatement of emissions
in response to the methane tax of 85 Mt/yr under
SSP1, only two thirds of that seen under SSP2.
Hence, unsurprisingly, the influence of the
baseline scenario becomes progressively more
important over time.
In addition to evaluating the emission impacts
of a global methane tax, the control cost of the
policy in each model year was also evaluated.
The GCAM reports the marginal cost of control,
which is equivalent to the tax. Many of the
resulting emission reductions, however, may
have been achieved for less. Therefore, to
estimate the policy cost in a particular year, five
additional scenarios were created with taxes
that were 10, 20, 40, 60, and 80 per cent of the
original methane tax trajectory. Integrating these
under the resulting curve for a particular year
produced an estimate of the control cost for that
year. Note that the control cost is not intended to
represent the overall policy cost, the calculation
of which would include economic feedbacks.
Analysis of the costs in GCAM5.2 show
that measures considered low cost, less
than US$ 600 per tonne of methane, in this
assessment can abate roughly 72 Mt/yr (Figure
4.16), which corresponds to about 16 per cent
of anthropogenic methane emissions under
SSP2. Under SSP1, the abatement is reduced
to ~54 Mt/yr, though this represents a fairly
similar 15 per cent of anthropogenic methane
emissions as those are reduced relative to SSP2
under that scenario’s baseline. These values are
fairly similar to the sum of the low-cost sectoral
abatement potential estimated in the analyses
by Harmsen et al., IIASA and the US EPA (Figure
4.10). The mitigation across sectors, however,
being heavily weighted towards abatement
within the fossil-fuel sector, is similar only to that
in the US EPA analysis and differs substantially
 Global Methane Assessment / METHANE EMISSIONS MITIGATION 112

from the other two. With the full tax implemented
in 2030, ~US$ 1 300 per tonne of methane,
roughly 21 per cent or 91 Mt/yr of emissions are
abated under SSP2, a value again fairly similar
to that found in the US EPA analysis (Figure
4.9). Within the fossil-fuel subsectors, however,
the GCAM’s abatement is overwhelmingly in
coal whereas the Harmsen et al., IIASA and the
US EPA analyses all show greater abatement
potential within oil and gas than within coal.
Overall, there is broad consistency in the
proportion of anthropogenic emissions that can
be abated by targeted measures in approximately
a decade between GCAM5.2 and the bottom-up
sectoral analyses of abatement potentials. There
are marked differences in the proportion within
each of the three main anthropogenic methane
emission sectors, and indeed the GCAM is
an outlier within IAMs in producing minimal
abatement within the agricultural and waste
sectors. In addition, the balance of abatement
potential across the three main fossil fuels is
quite different in GCAM5.2 with respect to the
abatement potential analyses of the Harmsen
et al., IIASA and the US EPA. Nonetheless, all
these analyses indicate that roughly 20 per cent
of anthropogenic emissions could be eliminated
through targeted control measures that cost
less than ~US$ 1 200 per tonne of methane,
a value less than a third the societal benefit of
~US$ 4 300 per tonne based on the evaluation
in Section 3.5.

600

500

400
METHANE (Mt/yr)

300

200

100

0
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

YEAR

SSP2 BASELINE SSP2 METHANE TAX SSP1 BASELINE SSP1 METHANE TAX

Figure 4.14 Anthropogenic methane emissions in Global Change Assessment Model 5.2, in baseline shared
socioeconomic pathways 1 and 2 simulations and in the same simulations but with an economy-wide methane tax
of 2018 US$ 2 100 per tonne of methane, increasing annually by 5 per cent to 2040, 2010–2100, million tonnes
600
Source: UNEP and CCAC

500
 Global Methane Assessment / METHANE EMISSIONS MITIGATION 113

200
FOSSIL FUELS 350
AGRICULTURE

300

150
250
METHANE (MT/YR)

METHANE (MT/YR)
200
100

150

100
50

50

0 0
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
YEAR YEAR

WASTE 15
BUILDING
90

80

12
70

60
METHANE (MT/YR)

METHANE (MT/YR)
9
50

40
6
30

20 3

10

0 0
2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

YEAR YEAR

SSP2 BASELINE SSP2 METHANE TAX SSP1 BASELINE SSP1 METHANE TAX

Figure 4.15 Methane emissions in Global Change Assessment Model 5.2 as in Figure 4.14 but by sector, 2010–
2100, million tonnes

Source: UNEP and CCAC

FOSSIL FUELS AGRICULTURE
200 350

300
Figure 4.16 Marginal
1400
150
250 abatement cost for
economy-wide methane
METHANE (MT/YR)

METHANE (MT/YR)

200
100
1200
reductions relative
150
to the percentage of
50
100 anthropogenic emissions
1000 50
abated in 2030, 2018
US$ per tonne of
0 0
methane.
MAC (20185/tCH4

800
Note: The
Global Change
600
Assessment Model
5.2 under a shared
400 socioeconomic
pathway 1 and 2
projections
200
Source: UNEP and
CCAC
0
0 10 20 30 40 50 60 70 80 90 100

Mt CH4 REDUCTION

SSP1 SSP2

1400
Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.4 METHANE MITIGATION AND
FOOD
Behavioural changes are an important component
of climate change mitigation according to
the IPCC (Mbow et al. 2019; Field et al. 2014),
though solutions presented in the literature are
overwhelmingly focused on their technological
aspects (van de Ven et al. 2017). Along with
options such as fuel switching and increased
efficiency discussed in the previous section,
demand management can play a substantial role
and may be especially relevant in the agricultural
sector in which targeted mitigation options leave
a large share of methane emissions unabated
(Sections 4.2 and 4.3). This section therefore
focuses on those options that are directly relevant
to methane emissions from the food supply chain.
Values quoted in this analysis are largely based
on the FAO’s GLEAM dataset (Gerber et al. 2013).
This includes methane emissions associated with
enteric fermentation and manure management,
as well as a small component from animal feed,
primarily for pigs and chickens.
4.4.1 REDUCING FOOD WASTE
Food loss and waste (FLW) is linked to
environmental, social and economic impacts
(FAO, 2019; Ishangulyyev et al. 2019; Mbow
et al. 2019; Abiad et al. 2018; FUSIONS 2016)
including substantial methane emissions.
According to the FAO (2019) food loss is defined
as loss occurring in the food supply chain from
harvest up to, but not including, the retail level,
while food waste is defined as loss occurring at
the retail and consumption levels (FAO, 2019).
There is no consensus to-date, however, on
separate definitions of food loss and food waste
(Ishangulyyev et al. 2019; Abdelradi et al. 2018).
Food that ends up in landfills with other municipal
solid waste is a source of methane.
While it occurs at various points along the supply
chain there are differences in food loss and waste
produced at different points between developing
and developed countries. In developing regions
most food waste occurs on-farm and during
distribution while in developed countries, food is
much more likely to be wasted because certain
foods are unappealing to consumers or because
they have passed their use-by dates (Porter et al.
2018; Garrone et al. 2014; Parfitt et al. 2010).
While more than 10 per cent of the global
population lives in hunger (FAO 2017) roughly a
12. Gigatonne = 109 tonnes
114
third of all food produced for human consumption
turns into lost or wasted at some point along the
food supply chain (Porter et al. 2018; Gustavsson
et al. 2011). Many studies highlight the mitigation
benefits of reducing this large volume and
indicate that the potential reductions of emissions
can be substantial but also diverse (FAO 2019;
Springmann et al. 2018; Wollenberg et al. 2015;
Bajželi et al. 2014). Most of these provide both
base case emissions and emissions reductions
estimates only in terms of carbon dioxide
equivalent rather than separating the various
greenhouse gases. For example, an FAO report
(2019) suggests that the global carbon footprint
of food loss and waste, excluding emissions from
land-use change, is 3.3 gigatonnes12 of carbon
dioxide equivalent (Gt CO2e). Similarly, an earlier
report from the FAO estimated total emissions
related to food loss and waste of 2.7 Gt CO2e
(FAO, 2014). Based on the source data reported
in Chapter 2, methane emissions from ruminants
and rice cultivation are ~145 Mt/yr. Hence if it is
assumed here that loss and waste in these two
categories is similar to the total across all food
types, methane emissions associated with food
loss and waste would be nearly 50 Mt/yr.
A prior study explored the implications of reducing
food loss and waste to reduce total agricultural
losses by 15 per cent by 2030, representing
roughly a 50–75 per cent decrease in food loss and
waste (Stehfest et al. 2013). Using two models,
the study found that this led to a reduction in
ruminant products of ~3–4 per cent and in crops
of ~6–9 per cent. These values are lower than the
theoretical maximum as prices decline for these
commodities with increased supply, leading to a
rebound effect that offsets some of the 15 per
cent reduction in food loss and waste. Such
decreases would lead to an abatement of ~5–7
Mt/yr of methane emissions. Larger reductions
could potentially be achieved with policies
focussed on preventing loss and waste of cattle
products, potentially including price adjustments
to prevent the rebound effects, as these have
the largest associated methane emissions.
Food loss and waste reductions would provide
additional benefits through reductions in carbon
dioxide and nitrous oxide emissions, though prior
valuation of the benefits of food loss and waste
reductions found that approximately two-thirds of
the benefits of such agricultural sector changes
are attributable to methane (Shindell et al. 2017) –
including the updates presented in Chapter 3 the
portion attributable to methane would be even
higher. Similarly, another study estimated a 2.2
Global Methane Assessment / METHANE EMISSIONS MITIGATION
per cent reduction in 2020 total greenhouse gas
emissions could be achieved by reducing food
waste (van de Ven et al. 2017). This translates to
about 1.3 Gt CO2e or about 40–50 per cent of the
food loss and waste total. Based on the 50 Mt/yr
of methane associated with food loss and waste,
this suggests reductions of as high as ~20 Mt/yr.
There are many potential strategies to reduce
food loss and waste for the different stages of
production, handling and storage, processing
and packaging, distribution and marketing, and
consumption. Table 4.2 presents a list of strategies
Table 4.2 Potential strategies for reductions in food loss and waste
DISTRIBUTION AND MARKETING
Improve inventory systems
Establish online marketplaces to facilitate sale
(donation) of perishable products
Change food date labelling practices and in-store
promotions
Improve institutions related to this stage
Improve distribution vehicles
Provide guidance on storage and preparation of food
to consumers
Improve the knowledge and ability of workers
Improve marketplaces – storage, covered areas
Deevelop links with research institutions to predict
consumer demand changes
Improve the cold-storage food chain
115
related to the distribution and marketing, and
consumption stages (Ishangulyyev et al. 2019).
Targeting behavioural change to women is
particularly important for the consumption stage
since women generally bear primary responsibility
for food purchase and preparation. Reductions
in food loss and waste can also come in part
from improvements in the cold food chain –
storage and transport – which not only influences
methane emissions but includes opportunities
to mitigate climate change by replacement of
hydrofluorocarbons with more climate friendly
substances.
CONSUMPTION
Facilitate increased donation of unsold foods from
cafeterias and restaurants
Implement consumer education and campaigns, both
nationally and regionally
Reduce portion sizes
Provide education about home economics in
educational institutions and communities
Involve women in food safe campaigns
Effective use of leftovers
Training for restaurant, cafeteria, and supermarket
management to forecast customer demand and reflect
demand in food purchasing to avoid bulk purchases
Implement good storage practices
Correctly interpret label dates
Distribution of excess food to charitable groups
Global Methane Assessment / METHANE EMISSIONS MITIGATION
4.4.2 IMPROVING LIVESTOCK MANAGEMENT
Farm management practices vary widely around
the world, and hence in regions with greater
emissions per unit of animal protein produced
there are substantial mitigation opportunities
through the adoption of best practices. This
issue was recently assessed by the FAO (Gerber
et al. 2013), which reported wide disparities in
emissions per kilogram of protein using the
Global Livestock Environmental Assessment
Model (GLEAM) framework. That dataset was
analyzed to explore the abatement potential if
the livestock production chain worldwide were
to match the practices of the most efficient
producers, where the latter were assessed
based on the 10th percentile of the lowest
emission intensity – i.e. a high-performing value,
but not an extreme case. In such an instance,
methane emissions from ruminants could be
reduced by ~31 Mt/yr. By far the largest share of
this abatement would come from cattle, which
contribute ~28 Mt/yr- – ~20 Mt/yr from meat
and ~8 Mt/yr from dairy. Another 5 Mt/yr might
be abated by the adoption of best management
practices for buffalo, which were not included
in this analysis due to a lack of data specifying
the 10th percentile of emissions per kilogram of
protein in the FAO dataset.
Achieving very low emissions per kilogram of
protein may involve large-scale industrialized
agriculture, which can have other social and
environmental impacts beyond greenhouse gas
emissions and hence such policies need to be
considered with care. It should also be noted that
improved livestock management is sometimes
considered in the targeted mitigation options
discussed previously (Section 4.2), though
clearly some of those analyses estimate much
lower potential for this as they find less total
mitigation potential across the entire agricultural
sector than that estimated here for livestock
management alone. In addition, changes to
livestock feed are commonly included in the
targeted analyses (Section 4.2) and constitute
another component of livestock management.
4.4.3 DIETARY CHANGE
According to the FAO (2019) between 2007–
2016 current food consumption patterns
were responsible for 21–37 per cent of
total greenhouse gas emissions. The FAO
furthermore estimates that the world will need
to produce about 50 percent more food by
2050 to feed the growing world population (FAO
2017). Unless current dietary patterns and food
116
systems change, this would result in significant
increases in greenhouse gas emissions (Drew et
al. 2020; FAO 2019; Poore and Nemecek 2018;
Alexandrowicz et al. 2016).
As described in Chapter 2, both rice cultivation
and especially ruminant animals are large
sources of methane emissions. It is therefore
not surprising that several studies suggest that
dietary shifts could significantly contribute to
reducing greenhouse gas emissions, including
methane (Clark et al. 2019; Springmann et al.
2018; Bryngelson et al. 2016; Springmann et al.
2016; Bajželj et al. 2014; Hedenus et al. 2014;
Pradhan et al. 2013).
Poor diet is also itself a major global health
concern, resulting in an estimated 11 million
premature deaths in 2017. Consequently, setting
dietary and nutritional goals would have great
benefits for human health through reducing
the incidence of non-communicable diseases
such as type 2 diabetes and cardiovascular
disease (Afhsin et al. 2019). It is important,
however, to mention that setting both dietary
and nutritional goals without considering the
environmental cost associated with them could
in fact result in an increase of greenhouse gas
emissions (Torero 2020). Recent studies also
demonstrate the multiple benefits of healthier
and more sustainable diets, including benefits
to human health and land use (Gao et al., 2018;
Springmann et al., 2016; Tilman et al. 2014;
Popp et al. 2010; Stehfest et al. 2009). The
IPCC Synthesis Report provides further detailed
analysis in relation to food systems, food
security, greenhouse gas emissions, and land
degradation (Pauchari et al. 2014).
There are three types of diets described by the
FAO in terms of the nutrients received by each diet:
1. energy sufficient diet: provides adequate
calories for energy balance for work each day;
2. nutrient adequate diet: provides adequate
calories and relevant nutrient intake values;
3. healthy diet: provides adequate calories
and nutrients, but also includes a more
diverse intake of foods from several
different food groups.
While each of these diets comes with a climate-
related cost, a critical point presented in the
FAO report (2020) is that not all healthy diets
are sustainable and not all diets designed for
sustainability are always healthy. The FAO
highlights, “this important nuance is not well
understood and is missing from ongoing
Global Methane Assessment / METHANE EMISSIONS MITIGATION
discussions and debates on the potential
contribution of healthy diets to environmental
sustainability” (FAO 2020).
Policies that integrate and combine
environmental and nutritional priorities are
limited (Aleksandrowicz et al. 2016; Watts et al.
2015). Furthermore, as is highlighted in Gao et al.
(2018), several studies suggest that combining
health, air pollution and greenhouse gas
benefits can lead to more attractive strategies
that encourage implementation of mitigation
measures related to dietary choices (Gao et al.
2018; Watts et al. 2017; Edenhofer et al. 2014;
Field et al. 2014; Haines et al. 2012; Shindell et
al. 2012; Chae et al. 2011; Watts et al. 2009;
Aunan et al. 2006; Cifuentes et al. 2001).
As with food waste reduction, most of the
studies of dietary change provide emissions
reductions estimates only in terms of carbon
dioxide equivalent rather than separating out the
various greenhouse gases. Several, however,
provide estimates of the change in consumption
of cattle-based foods, for which the impact on
methane emissions can readily be evaluated.
For example, a study of diets in the United
Kingdom explored how alterations could both
optimize nutrition and maximize greenhouse
gas reductions without drastically changing
overall dietary patterns. This study found that
a reduction of 34 per cent in the consumption
of both beef and dairy products (average for
men and women) was ideal (Green et al. 2015).
If extended across the European Union, and
assuming a similar percentage decrease was
ideal across that region, this would lead to a
reduction in methane emissions of ~6 Mt/yr
based on the full lifecycle emissions associated
with livestock (Weiss and Leip 2012).
In a broader study, it was estimated that the
adoption of healthy diets globally would entail
approximately a 50 per cent reduction in beef
consumption in 2050 (Stehfest et al. 2009).
Accomplishing such as shift in the near term
would lead to a reduction in methane emissions
from beef cattle of ~14 Mt/yr based on current
consumption. As demand for red meat is
projected to roughly double by the middle of
this century, the overall benefits in the future
would be larger, though obviously they would
depend on a reduction that could be achieved
by a target year such as 2030. Including such a
dietary shift in climate change mitigation policies
is also estimated to greatly reduce the costs of
achieving low-warming targets (Stehfest et al.
2009). These results seem generally comparable
to the overall greenhouse gas reductions
117
estimated for dietary change, which are in the
range of 4.4 per cent for healthy diets and ~6–7
per cent for vegetarian or vegan diets in 2030 (van
de Ven et al. 2017). Given that livestock accounts
for roughly 18 per cent of total greenhouse gas
emissions (Stehfest et al. 2009), the healthy diet
reductions translate to a decrease in livestock-
related emissions of about 25 per cent. Based
on total ruminant emissions of 115 Mt/yr of
methane (Chapter 2), this corresponds to diet-
related reductions of ~29 Mt/yr of methane if the
reductions are comparable across consumption
of the various potential animal products.
These estimates provide a rough idea of the
emissions reductions that could be achieved
by a reduction in the consumption of ruminant-
based foods, primarily from cattle through the
adoption of either healthy or vegetarian/vegan
diets. Reductions associated with healthy diets
are likely to represent the majority of those that
would come from vegetarian diets, with lower
barriers to adoption. Healthy diets might achieve
reductions in methane emissions in the range
of 15–30 Mt/yr, with additional climate benefits
from reductions in carbon dioxide and nitrous
oxide – the IPCC, for example, estimates a total
mitigation potential from healthier and more
sustainable diets of 0.7-8.0 Gt CO2e/yr by 2050
(Mbow et al. 2019).
4.4.4 CONCLUSION ON METHANE
MITIGATION AND FOOD
Additional possibilities for mitigation of the large
methane emissions from agriculture, especially
livestock, exist and some cannot be easily
characterized as solely targeted or additional.
For instance, substitution of cultured meat for
traditional livestock products could substantially
reduce associated greenhouse gas emissions,
especially methane (Post 2012; Tuomisto and
Teixeira de Mattos 2011). Such technologies are
not yet commercially viable, however, and may
not gain widespread consumer acceptance.
Overall, while individual behaviour cannot
address the scale of greenhouse gas emissions,
literature suggests that including behavioural
change in a broad portfolio of policies could
help address and mitigate emissions and also
support top-down policies (IPCC 2018; van de
Ven et al. 2017; Watts et al. 2015; Gilligan et al.
2010). Implementing these changes, however,
may be challenging for several reasons. There
will be a clear discrepancy in the priorities, needs
and abilities of different national governments to
assist with these changes. Taking the example
Global Methane Assessment / METHANE EMISSIONS MITIGATION
of diets, many developed countries have an
obesity epidemic, while in some developing
ones, hunger and malnutrition, through having
too little or not enough of the right types of food,
is a severe problem. Switching diet to more
sustainable sources will have co-benefits for
reducing greenhouse gas emissions, particularly
methane, and improving health (Willet et al.
2019). Implement structural and long-lasting
changes in individual dietary intake, however,
will likely require strong intervention, mitigation
and incentivisation by governments through
innovative policies. Social influence approaches
can be an effective way of encouraging resource
conservation (Camilleri et al. 2019; Abrahamse
and Steg 2013), as part of either government
or non-governmental programmes. Similarly,
interventions in livestock management practices
or efforts to reduce food loss and waste face
considerable barriers in terms of institutional
capacity, education and societal reluctance to
change traditional habits as well as economic
barriers. Nevertheless, given that the impact of
the three elements with the strongest evidence
base – reduced food loss and waste, improved
livestock management, and change to healthier
diets – is estimated here to have the potential
to reduce methane emissions by up to 65–80
Mt/yr over the next few decades, the impact of
behavioural change and innovative policies on
agricultural methane emissions should not be
ignored. The full abatement potential identified
here for these elements is unlikely to be realized
by 2030, so that abatement at these levels is
perhaps an appropriate target for 2050.
4.5 WEB-BASED DECISION SUPPORT
TOOL
A web-based decision support tool that allows
users to input a level of methane emissions
mitigation and easily determine the national
values for the response of all impacts included
in this analysis has been developed. This can
be used to evaluate the benefits to climate,
health and, through environmental changes, the
economy of a wide range of action, ranging from
individual mitigation projects to national pledges
118
under the Paris Agreement or international
targets for methane controls.
The tool also allows the user to view the impacts
of methane mitigation options based on the four
mitigation potential analyses described in this
assessment – Harmsen et al., IEA, IIASA, and
the US EPA. The user can select broad sectors
or more detailed subsectors and can choose to
view impacts of worldwide or regional mitigation.
For example, a user wanting to examine
mitigation potential of methane emissions from
the fossil-fuel sector based on the IIASA analysis,
can select low-cost measures only, and the
global total, and get the impact on respiratory
health from the associated methane emissions
reductions. By placing the cursor over a country
a panel will pop up showing the reduction in
national-level deaths per million persons and the
numerical value for that country’s results will be
displayed. Worldwide totals and the associated
mitigation costs and benefits will be displayed
below the interactive maps.
Note that the temperature responses in the tool
include the estimated carbon cycle response
enhancement of 10 per cent (Section 3.3),
though this has not been applied to heat-
related labour productivity changes, for which a
10 per cent change would be far less than the
associated 70 per cent uncertainty. This tool can
be found at: http://shindellgroup.rc.duke.edu/
apps/methane/
An archive of the hourly ozone produced by the
modelling described in Chapter 3 has also been
created. This will allow calculation of any new
health or agriculture metrics should the finding
for which metric is most appropriate change
again. This repository is available at https://doi.
org/10.7924/r4qn65b0z and can be cited as:
Shindell, D., Zhang, Y., Seltzer, K., Faluvegi,
G., Naik, V., Horowitz, L., He, J., Lamarque,
J.-F., Sudo, K., and Collins, B. 2020. Data
from modelling in support of the Global
Methane Assessment, UN Environment and
Duke Research Data Repository. https://doi.
org/10.7924/r4qn65b0z
5. CONCLUSIONS:
MITIGATION
OPPORTUNITIES,
IMPACTS AND
UNCERTAINTIES
CHAPTER FINDINGS
• Implementation of identified targeted control measures with costs less than
the estimated societal benefits could abate ~100 Mt/yr of methane emissions
by 2030 and provide benefits including reducing global warming by ~0.15ºC
over the 2040–2070 period, and by ~0.2º C over the longer term (2070–2100);
preventing premature deaths due to ozone exposure immediately with
annual totals reaching ~140 000 per year in 2030; preventing ~40 billion lost
work hours annually due to heat exposure annually by ~2045; and preventing
~15 Mt tonnes of crop losses due primarily to ozone exposure each year by
2030. The crop yield increases would be worth approximately US$ 4 billion
per year, the increased labour productivity about US$  6.3 billion per year,
and the valuation of the reduced risk of premature deaths is ~US$ 250 billion
(US$60–420; 95 per cent confidence interval).

• T
 he greatest targeted abatement potential very likely lies within the
fossil-fuel sector, for which average mitigation costs range from a high of
~US$1 250 per tonne of methane in the US EPA analysis to a net savings of
~US$ 400 per tonne in the IIASA analysis, with values in between those for
the Harmsen and IEA analyses, the latter of which only considers the oil and
gas subsectors. All these average costs are much lower than the per tonne
societal benefits.

• E
 xamining only targeted controls in the relatively robustly characterized
fossil-fuel and waste sectors that cost less than the estimated societal
benefits per tonne, the abatement potential is about 75 ±2 Mt/yr. Benefits
from this level of abatement include ~0.13ºC warming avoided in the 2040–
2070 period, ~105  000 avoided premature deaths due to ozone exposure
annually by 2030, and 11 Mt of avoided crop losses due primarily to ozone
exposure each year. The benefits of the annually avoided premature deaths
alone are estimated at US$ 190 billion (US$  40–310 billion; 95 per cent
confidence interval).

• T
 argeted measures alone are not enough to reach emissions levels seen in
1.5ºC scenarios over the next several decades, which require both targeted
and additional measures to achieve ~180 Mt/yr of methane abatement
by 2030. Behavioural measures include fuel switching, energy efficiency,
improved waste separation, reducing food waste, dietary change and
transport demand management. Both targeted and additional measures
face implementation barriers including financing, a lack of awareness/
education, governance and consumer preferences.

• U
 ncertainties exist along the entire chain from mitigation potential through
physical responses to societal impacts. Overall uncertainties are likely to
be primarily driven by limited understanding of climate sensitivity and of the
effects of ozone exposure on human health, with generally small contributions
from incomplete knowledge of mitigation potentials or in an understanding
of the ozone response to methane emissions reductions. In some specific
sectors or regions, however, limited knowledge of current emissions is
probably a major driver of uncertainties in benefit-potential estimates.
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
It has been clear to the scientific community for
some time that reducing methane emissions
would contribute to both improved air quality
and climate change mitigation (Shindell et al.
2012; UNEP 2011; UNEP/WMO 2011; Fiore et
al. 2002). There are substantial uncertainties,
however, in the characterization of each step
along the pathway from mitigation to impacts,
including the methane emissions themselves,
the responses of the physical system to
changes in emissions, and the potential for
emissions reductions in the various sectors
that emit large amounts of methane. Prior
analyses have typical relied on the results of
one or two models for physical responses,
used a single set of emissions data and a single
set of mitigation potentials. This assessment,
however, provides a more thorough analysis of
robustness for each step by analyzing results
from multiple models of the composition-
climate system, multiple models of the energy-
economy-land system, and multiple analyses
of methane mitigation potentials. It also
incorporates recent advances in understanding
of the societal impacts of physical system
responses to methane emissions, including
not only climate and composition responses
of the physical system, but many downstream
impacts of both ozone and climate change on
human health and crops. This provides a robust
characterization of the overall societal costs
and benefits of methane mitigation.
The analysis shows that there is great potential
to mitigate methane emissions over the current
decade, 2020–2030, using existing targeted
control measures. Such measures could abate
roughly 125 Mt/yr of methane emissions, about
30 per cent of the projected 2030 anthropogenic
emissions. Taking the average across the various
available analyses of abatement potentials and
costs, targeted control measures with costs
of less than the estimated societal benefits
could abate 101 Mt/yr of methane emissions
by 2030 and achieving such abatement would
have enormous societal benefits. These include
climate-related benefits of reducing warming by
~0.15ºC by 2040, with a value of 0.2ºC over the
longer term (~2070–2100), and the associated
prevention of ~40 billion lost work hours, roughly
30 per cent of which are for women and 70 per
cent for men, due to heat exposure annually,
starting around 2040. They also include air
pollution-related benefits starting immediately
after emissions are reduced and continuing each
year thereafter. Increasing through the 2020s,
benefits for 2030 and subsequent years include
preventing annually ~140 000 premature deaths
121
due to ozone exposure, ~430 000 accident and
emergency department visits due to asthma
and ~9 000 hospitalizations of elderly people
due to ozone, and ~15 Mt of crop losses due
to ozone exposure. The crop yield increases
would be worth approximately US$ 4 billion per
year, the increased labour productivity about
US$ 6.4 billion per year, while the valuation
of the reduced risk of premature deaths is
~US$ 250 billion annually (US$ 60–420; 95 per
cent confidence interval).
The discussion of climate change responses to
methane mitigation in Chapter 3 focussed on
the response averaged over 1–3 decades after
potential cuts. It is useful to also characterize
the potential impact of methane mitigation on
global mean temperature as a function of time.
Therefore the response was evaluated using
absolute global temperature potentials (AGTPs),
as in prior Assessments (UNEP 2017; UNEP/
WMO 2011). In brief, the yearly AGTPs represent
the global mean temperature change per
kilogram of emission each year subsequent to
those emissions based on an impulse-response
function for the climate system, as is used in
IPCC for selected example years, for example,
AGTP50 or AGTP100 (Myhre et al. 2013). In the
calculations in this assessment, the transient
climate response is based on analysis of the last
generation of climate models (CMIP5) (Geoffroy
et al. 2013). The AGTPs include the carbon-cycle
response to the temperature change induced
by the emitted species including the impact of
ozone generated by methane on carbon uptake
(Gasser et al. 2017; Collins et al. 2010) as
described in Shindell et al. (2017). Temperature
responses are also provided averaged over
latitude bands based on analogous absolute
regional temperature potentials (Collins et al.
2013; Shindell 2012). To take advantage of
the new results from simulations in support
of this assessment, the responses have been
calibrated to match the global climate impacts
reported in Chapter 3, including the estimated
carbon cycle response enhancement of 10 per
cent (Section 3.3), though this changes the
AGTPs only minimally.
To illustrate the evolution of temperature
responses over time, the temperature response
to the methane mitigation seen in 1.5ºC
scenarios is evaluated, based on IAMs that
included targeted controls, behavioural change,
emissions pricing and other developments such
as decarbonization. The average mitigation
across the available scenarios is 180 Mt/yr in
2030, 240 Mt/yr in 2040 and 280 Mt/yr in 2050.
Avoided warming reaches 0.3ºC towards the
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
end of the 2040s in response to those emissions
reductions (Figure 5.1). In comparison with prior
studies, the 2050 value of 0.32ºC is slightly greater
than the impact of methane mitigation estimated
in Shindell et al. (2012)/ UNEP/WMO (2011)
for 2050, 0.28ºC, but is substantially so when
comparing both results for period 30 years after
abatement begins – 2040 in the earlier scenarios,
for which time the response was 0.24ºC in
response to abatement of 139 Mt yr in 2030. The
greater 2050 value in this analysis stems primarily
from the greater abatement amounts which have
a larger impact than the later start of abatement.
The Arctic response is 60 per cent larger than
the global mean value, in good agreement with
the results found here in the full climate model
simulations (Chapter 3; Table 3.2).
Comparisons are also drawn with the 2017
UNEP Emissions Gap Report that estimated
that the response to the full application of
targeted methane mitigation options identified
by IIASA would lead to avoided warming of
0.09ºC in 2030 and 0.30±0.12ºC in 2050 using
a similar method to that used here (UNEP 2017).
Methane mitigation analyzed in that report was
152 Mt/yr by 2030 and 171 Mt/yr in 2050. Those
are greater values for 2030 than reported here
from the IIASA bottom-up analyses as the older
ones used in the UNEP report had an earlier
abatement starting date. The results presented
here exhibit a slightly smaller response per million
tonnes of emissions mitigation based upon the
latest climate modelling, though well within the
0.0
-0.1
-0.2
-0.3
°C
-0.4
-0.5
-0.6
2020 2025 2030
SOUTHERN HEMISPHERE EXTRATROPICS
NORTHERN HEMISPHERE MID-LATITUDES
Figure 5.1. Temperature response to methane abatement from 2020–2050 based on mitigation levels consistent
with 1.5º C scenarios, 2020–2050, degrees centigrade
Note: In addition to global mean responses, values are given for the southern hemisphere extratropics
(90–28º S), the tropics (28º S–28º N), the northern hemisphere mid-latitudes (28–60º N) and the Arctic
0.0
(60–90º N).
Source: UNEP and CCAC
-0.1
122
uncertainty range. Of note is that, based on both
the new modelling and updated constraints on
climate sensitivity, the uncertainty ranges are
slightly reduced in the new analyses: 95 per
cent confidence intervals were 40 per cent in the
2017 UNEP report and 36 per cent in Shindell et
al. (2012)/UNEP/WMO (2011), whereas now they
are estimated at 33 per cent (Chapter 3).
In addition to this analysis of the impact of
emissions changes through 2050, the impact
of methane emissions changes through 2100
were also explored, again using AGTP. In 1.5º
C scenarios generated by IAMs, methane
emissions fall rapidly in the first half of the
century but thereafter decline only modestly
through 2100 (Rogelj et al. 2018). As discussed
in Chapter 4, this is primarily due to emissions
from agriculture which are not abated given the
cost curves, abatement options and pricing
mechanisms used in most of the IAMs. As such,
the longer-term climate impact of aggressive
methane abatement becomes highly dependent
upon the reference case against which the fairly
steady post-2050 are compared (Figure 4.1).
Relative to the high baseline methane emissions
under SSP3’s marker scenario, the 2100 impact
of 1.5º C methane reductions is a reduction
in warming of 0.77º C. Relative to the lower
baseline emissions of SSP1’s marker scenario,
the reduction is much less at -0.42º C, with the
mid-range (for methane) SSP4 and SSP5 having
value of about -0.6º C.
2035 2040 2045 2050
GLOBAL TROPICS
ARCTIC
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
The identified targeted abatement measures
reduce emissions in the fossil fuel sector by
~58 ±12 Mt/yr and those in the waste sector
by ~32 ±4 Mt/yr when examining all controls
regardless of cost for 2030. The targeted
abatement potential within the agricultural
sector is not as robustly characterized as the
estimates are very different at ~10, 20 and 50 Mt/
yr. Hence this analysis reveals that the greatest
targeted potential very likely lies within the
fossil-fuel sector, whereas there is the greatest
agreement on the abatement potential in the
waste sector. Delving further into the emission
sources, this assessment finds that within
the fossil-fuel sector there is a comparable
magnitude abatement potential in each the
three subsectors coal, gas and oil. In contrast,
within the waste sector the abatement potential
is largely concentrated within the solid waste
subsector with a relatively small amount within
the wastewater subsector. These results consider
all abatement measures regardless of cost and
show substantial differences in the average
value, weighted by mitigation potential at each
cost value, across the various analyses. For the
fossil-fuel sector, average mitigation costs range
from a high of US$1 240 in the US EPA analysis
for measures costing less than US$ 25 000 per
tonne of methane), a middle value of US$ 960 per
tonne of methane in the Harmsen et al. analysis
to -US$ 400 in the IIASA analysis, with values in
between those from IEA, US$ 10 for the oil and gas
subsectors only. Although the mitigation potential
for the waste sector is more consistent across the
various analyses, the costs are more divergent,
with an average cost range of US$ 3 930 per
tonne of methane in the Harmsen et al. analysis,
$840 in the US EPA analysis for measures costing
less than US$ 25 000 per tonne of methane, to
-US$ 5 800 in the IIASA analysis. In comparison,
average costs for the agricultural sector are more
consistent at US$ 1 390 in Harmsen et al., US$
800 in the US EPA analysis and US$ 270 in IIASA,
despite the large divergence in the abatement
potential itself for this sector. As discussed in
Section 4.2.1, abatement potentials vary greatly
across regions. The largest potential in the waste
sector is for Europe and India; the coal subsector
followed by livestock for China; in livestock
for Africa, followed by oil and gas, in the coal
subsector and the waste sector for Asia and the
Pacific (excluding China and India); in the oil and
gas subsector for the Middle East, North America,
and Russia and the Former Soviet Union; and in
the livestock subsector for Latin America.
123
The targeted mitigation potential was examined
for only those measures with costs of less
than the US$ 4 300 per tonne of methane in
estimated societal benefits found in Chapter
3. The total abatement potential in the three
cross-sector analyses are 85 Mt/yr in IIASA, 93
Mt/yr in US EPA and 121 Mt/yr in Harmsen et
al., for an average of 100 Mt/yr. In that same
order, potentials are 46, 44, and 59 Mt/yr for
the fossil-fuel sector, 27, 30 and 18 Mt/y for the
waste sector, and 12, 19 and 44 Mt/yr for the
agricultural sector. Hence the totals across the
more robust fossil-fuel and waste sectors are 73,
74 and 77 Mt/yr, for a range of 75 ±2 Mt/yr. The
benefits from this level of abatement (75 Mt/yr)
are also substantial: just over 0.1ºC of warming
avoided a decade after these reductions are
achieved, and benefits immediately following
reductions, and continuing each year thereafter,
that include about 105 000 avoided premature
deaths due to ozone exposure, about 325 000
fewer accident and emergency department visits
due to asthma, and ~11 Mt of crop losses due
to avoided ozone exposure. The benefits of the
avoided premature deaths alone are estimated
at US$ 190 billion per year. It is emphasized that
these benefits all come at a net gain to society
as even the most expensive controls within this
analysis have costs of less than the societal
benefits, and hence the average costs of controls
are far less than the average benefit per tonne of
emissions abatement.
It is also informative to examine the total cost
of implementing all targeted methane measures.
The IEA analysis is excluded here as it does not
encompass the full fossil-fuel sector, and the
IIASA and US EPA analyses are relied on since the
Harmen et al. analysis does not include negative
costs. For the fossil fuel sector, the total cost
based on the US EPA analysis is US$ 62 billion
for measures costing less than US$ 25 000
per tonne of methane, whereas the total is a
savings of US$ 18 billion in the IIASA analysis.
Similarly for waste, the total cost based on the
US EPA analysis is US$ 27 billion whereas the
total is a saving of US$ 166 billion in the IIASA
analysis. The differences in both sectors are
partly attributable to the much higher discount
rate used in the US EPA analysis –10 per cent
versus 4 per cent in the IIASA analysis – which
greatly reduces the value of gas that is captured
and utilized – net costs will also depend on
the estimated future price of natural gas. In
the waste sector, the inclusion of substantial
revenue from recycled paper in the IIASA waste
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
analysis, as discussed in Chapter 4, is another
major source of differences. Hence the total
cost for all targeted measures across these two
sectors ranges from ~US$ 90 billion to a savings
of ~US$ 185 billion. For the agricultural sector,
the total cost based on the IIASA analysis is
US$ 3 billion whereas the total is US$ 17 billion
in the US EPA analysis. Thus the spending in
the agricultural sector is more consistent, even
though the mitigation potential differs strongly
– see Chapter 4 for further discussion on the
agricultural sector in these two analyses. This
is likely due to the reduced influence of the
discount rate differences as less of the methane
abatement is used in this sector. Hence it is very
difficult to estimate the total spending required
to implement the targeted measures. Averaging
across these two analyses for all three sectors,
the total spending is negative. In all sectors the
total cost seems likely to be relatively small or
negative but will depend on the chosen discount
rate and prices of natural gas. As the total costs
include both measures with net savings and
those with net costs, sometimes large costs,
it is important to keep in mind that even when
total costs are large, they are sensitive to the
costs of the most expensive measures along
the cost curves. Thus the conclusions regarding
the portion of targeted measures available at
negative or low costs are typically emphasized
in this assessment.
The uncertainty in the fossil-fuel plus waste sector,
the methane mitigation potential is about 17–20
per cent, though it is larger for some subsectors
within those general sectors. Considering the all-
cost measures, the uncertainty is about 25 per
cent for the total from fossil fuels and about 10
per cent for the total from waste. The variation
in these mitigation potentials can be compared
with the uncertainties in climate, health and crop
responses. For climate, there is about a 34 per
cent uncertainty in the global mean temperature
change, based on current understanding of
climate sensitivity, with a slightly smaller value
for the tropics and a larger one for the Arctic,
here 60–90º N, where the response is largest.
For premature deaths, uncertainties for those
related to respiratory effects are ~35 per cent
and those for cardiovascular-related effects are
~65 per cent, both for the 95 per cent
confidence interval. Uncertainties in heat-
related deaths, for China and the United States
only, are about 48 per cent with a 95 per cent
confidence interval, whereas those for crop
yield impacts are roughly 36 per cent. These
impact uncertainties reflect the fact that the
temperature response to methane emissions
124
changes, ~34 per cent, is more uncertain than
the ozone response, ~24 per cent. In general, it
was found that uncertainties in the impacts of
methane mitigation are likely to be driven most
strongly by uncertainties in climate sensitivity
and in the effects of ozone exposure on human
health, with relatively small contributions from
uncertainties in the mitigation potential or ozone
response to methane emissions reductions. It is
also noted that there likely are additional effects
of ozone on human health, as recent research
indicates that exposure to air pollution affects
nearly every organ in the human body (Andrade
et al., 2019), and hence additional impacts are
likely to be present beyond those quantified
here, so that health benefits of mitigation should
be considered conservative.
Considering the low-cost measures, defined
here as those costing less than US$ 600 per
tonne of methane, the different analyses vary
more as they are sensitive to the use of a cost
threshold at this level since it falls within a range
encompassing many mitigation measures. For
mitigation within the fossil-fuel sector, low-cost
mitigation potentials range from 32 to 66 Mt/yr
in the two analyses finding that the average cost
of the low-cost measures is negative, -US$ 120
and -US$ 851 per tonne of methane in the IEA
and IIASA analyses, respectively. The Harmsen
et al. and US EPA analyses, however, find low-
cost potentials of 24 and 42 Mt/yr and average
costs that are positive at US$ 182 and US$
38 per tonne of methane, respectively. For the
waste and agricultural sectors, the average costs
of the low-cost measures is negative in some
analyses and very low in others, always less than
US$ 140 per tonne of methane and typically less
than US$ 60 per tonne of methane. Magnitudes
vary by nearly a factor of two in the waste
sector, 11–21 Mt/yr, and nearly a factor of four
in the agricultural sector, 10 to 38 Mt/yr. Hence
the low-cost measures are generally beneficial
from an economic standpoint even without
accounting for the environmental benefits, or
when only accounting for market costs, primarily
crop yield and labour productivity increases,
when evaluating environmental impacts, but
the potential available at these low costs is
quite sensitive to the particular analysis used.
As discussed previously, when environmental
benefits are accounted for, slightly more than
80 per cent of identified targeted potential for
methane controls lead to net societal benefits.
Though significant, a reduction of either ~100
or ~125 Mt/yr, which represents ~25 or 30 per
cent in anthropogenic methane emissions by
2030 is less than that seen in 1.5ºC scenarios.
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
Examining estimated targeted mitigation
potential over a longer timescale provides
valuable insights into the potential for targeted
controls to provide abatement consistent
with low warming scenarios. Looking at the
total assessed mitigation potential across
all sectors, IIASA concluded that maximum
feasible reductions would reduce 2050 methane
emissions by 242 Mt/yr whereas Harmsen
et al. found a 2050 total abatement potential
of 185 Mt/yr. These analyses account for
the time required for technological turnover
and for phasing in new regulations, and also
include changes in baseline emissions based
on projected energy demand and population
growth, which affects generation of waste and
food production. Baseline 2050 anthropogenic
emissions were projected to be 450 Mt/yr in the
IIASA analysis, a large increase over the 355 Mt/
yr in 2020 in its model. The reduction of 242 Mt/
yr thus represents a decrease of 54 per cent. The
reduction is similar in the Harmsen et al. analysis
at 47 per cent despite the smaller abatement as
that analysis had a smaller projected growth
in baseline methane emissions. An equivalent
percentage reduction today would translate to
about 190 Mt/yr of methane mitigation. That
level of reduction would be almost exactly
equal to the mean methane mitigation in 2030
seen in 1.5ºC scenarios. On the one hand, this
indicates that full implementation of targeted
mitigation measures in a single decade
could potentially lead to methane reductions
consistent with the magnitude required in
2030 for such a low warming pathway. On the
other hand, this indicates that were targeted
measures to be the only policies employed, full
application of all technologies, even those with
very high mitigation costs, would be required at
an exceptionally fast rate of deployment and that
even with targeted measures across the full cost
range, the additional methane mitigation after
2030 needed to stay on a 1.5ºC pathway would
require additional measures which contribute to
other priority development goals with benefits
for methane. The additional measures examined
in this assessment include several that might be
widely considered part of climate policy, such
as switching to renewable energy, and others
that might not, such as reducing food waste or
improving waste separation. Such categorizations
are highly subjective, but in general the additional
measures in the fossil-fuel sector are more
closely aligned with policies commonly viewed
as part of a climate portfolio. Those constitute
roughly 40 per cent of the additional measures
included in this assessment. Roughly 43 per cent
come from changes in the agricultural sector and
125
17 per cent from the waste sector, both of which
include measures associated with consumption
or disposal practices that are less consistently
currently viewed as part of climate policies.
Behavioural changes have the potential to
substantially augment targeted methane controls.
These are especially important for the agricultural
sector, where targeted measures have the smallest
relative impact. The analysis presented here
suggests that behavioural changes, reducing food
loss and waste, improving livestock management
and shifting to healthier diets that at the global level
include reduced consumption of ruminant-based
foods, have the potential to provide up to 65–80
Mt/yr of additional methane emissions mitigation.
Though these face substantial structural barriers,
widespread adoption of such measures alongside
targeted controls could bring anthropogenic
methane emissions in line with those in 1.5ºC
scenarios. Additional progress in targeted
measures including feed supplements for livestock
and alternatives to animal protein (Section 4.4.4)
as well as diverting organic material that would
otherwise lead to methane release from landfills
to make useful products also have potential if
they prove economically feasible and societally
acceptable, and hence can be scaled up.
Policy ambition to date has thus far fallen short of
mitigation consistent with 1.5ºC scenarios, and,
as discussed in Chapter 1, atmospheric methane
amounts are continuing to increase. Some steps
have been taken, however. The countries of North
America pledged to reduced methane emissions
from oil and gas by 45 per cent by 2025, and the
Climate and Clean Air Coalition (CCAC) Mineral
Methane Initiative is aiming for a reduction of
60–75 per cent from the oil and gas subsector by
2030 among participants. These would provide
substantial sector and region-specific mitigation.
For example, were the low end of the 2030 Oil
and Gas Initiative goal to be achieved across all
CCAC partner countries, that would correspond to
mitigation of about 30 Mt/yr of methane emissions
based on projected 2030 baseline emissions, or ~28
Mt/yr based upon current 2020 emissions. Several
sub-national jurisdictions have also put ambitious
methane reduction laws into place. These include
the US states of California and New York, with a 40
per cent reduction from 2013 methane emissions
by 2030 in the former and a 40 per cent reduction
from 1990 emissions by 2030 in the latter for all
greenhouse gases including methane – using
a 20-year methane global warming potential to
emphasize the importance of reducing this gas
as outlined in the New York’s Methane Reduction
Plan. Such steps are, however, only a partial recipe
for achieving a 1.5ºC pathway.
Global Methane Assessment / CONCLUSIONS: MITIGATION OPPORTUNITIES, IMPACTS AND UNCERTAINTIES
Achieving a 1.5ºC pathway would require
not only reversing the current rising trend in
methane emissions but also greatly reducing
anthropogenic methane emissions by 2030.
This could be accomplished by policies targeted
at implementing targeted controls on specific
sectors, policies to bring about behavioural
change, policies to bring about broad, cross-
sectoral emissions reductions such as ambitious
international agreements, following, for
example, the pattern of the Kigali Amendments
to reduce hydrofluorocarbon emissions,
imposition of emissions caps or taxes, such as
in the modelling presented in Section 4.3, or
a combination of such policies. Implemented
policies should be gender responsive, especially
those promoting behavioural change. Achieving
methane reductions also requires increased
education and information about the economic
and societal impacts of mitigation, something
this assessment aims to provide, along with
additional infrastructure to make use of abated
emissions in some sectors and a change in
economic incentives that may be misaligned
and prevent even action with overall negative
costs from being taken up (IEA 2020). It is also
worth pointing out that due to the removal
of co-emitted aerosols that accompanies a
realistically paced phase-out of fossil fuels, even
aggressive decarbonization will tend to have
126
minimal net impacts on near-term, the first 20
years, temperatures. (Shindell and Smith, 2019).
Prior research has therefore pointed out that
the only plausible way of decreasing warming
relative to the reference case during the next 20
years is through reductions in greenhouse gases
from non-fossil sources including methane and
nitrous oxide from agriculture and waste as well
as fluorinated gases (Shindell and Smith 2019).
Prior analyses have shown the multiple benefits
of reducing methane emissions (Shindell et al.
2012; UNEP/WMO 2011). This work adds to
those prior analyses by both including more
modelling to better characterize robustness
and analyses of additional impacts, and also by
greatly extending the scope of the associated
economic analysis. A prior economic analysis
by UNEP (UNEP 2011) included results from
a single analysis of mitigation costs, whereas
this study includes multiple such analyses and
finds that the variation across them of mitigation
potentials is large, leading to greater uncertainty
than in the analyses of physical system
responses, especially for the agricultural sector.
Results are relatively robust for the fossil fuel
and waste sectors and indicate clearly that there
is an enormous potential for methane emissions
mitigation with benefits that greatly outweigh
implementation costs.
6. REFERENCES
Global Methane Assessment / REFERENCES 128

Abbott, D.W., Aasen, I.M., Beauchemin, K.A., Grondahl, F., Gruninger, R., Hayes, M., Huws, S.,
Kenny, D.A., Krizsan, S.J., Kirwan, S.F., Lind, V., Meyer, U., Ramin, M., Theodoridou, K., von Soosten,
D., Walsh, P.J., Waters, S., and Xing, X. (2020) Seaweed and Seaweed Bioactives for Mitigation of
Enteric Methane: Challenges and Opportunities. Animals, 10(12), 2432.

Abdelradi, F. (2018). Food waste behaviour at the household level: A conceptual framework. Waste
Manag., 71, 485-493.

Abiad M.G. and Meho L.I. (2018). Food loss and food waste research in the Arab world: A systematic
review. Food Security, 1–12.

Abrahamse, W. and Steg, L. (2013). Social influence approaches to encourage resource conservation:
A meta-analysis. Global Environ. Change, 23, 1773–85.

Academy of Science of South Africa, Brazilian Academy of Sciences, German National Academy of
Sciences Leopoldina, U.S. National Academy of Medicine and U.S. National Academy of Sciences.
(2019). Air Pollution and Health – A Science-Policy Initiative. Annals of Global Health, 85, 140, 1-9.

Afhsin, A., et al. (2017). Health effects of dietary risks in 195 countries, 1990–2017: a systematic
analysis for the Global Burden of Disease Study. The Lancet, 393, 1958-1972.

Alexandrowicz, L., Green, R., Joy. E.J.M., Smith, P. and Haines, A. (2016). The Impacts of Dietary
Change on Greenhouse Gas Emissions, Land Use, Water Use, and Health: A Systematic Review.
PloS ONE, 11(11): e0165797.

Alexe, M., Bergamaschi, P., Segers, A., Detmers, R., Butz, A., Hasekamp, O., Guerlet, S., Parker, R.,
Boesch, H., Frankenberg, C., Scheepmaker, R.A., Dlugokencky, E., Sweeney, C., Wofsy, S.C., and
Kort, E.A. (2015). Inverse modelling of CH4 emissions for 2010–2011 using different satellite retrieval
products from GOSAT and SCIAMACHY. Atmos. Chem. Phys., 15, 113–133, doi:10.5194/acp-15-
113-2015.

Alvarez, R.A., Zavala-Araiza, D., Lyon, D.R., Allen, D.T., Barkley, Z.R., Brandt, A.R., Davis, K.J.,
Herndon, S.C., Jacob, D.J., Karion, A. and Kort, E.A. (2018). Assessment of methane emissions from
the US oil and gas supply chain. Science, 361, 186-188.

Anderson, G.B., Oleson, K.W., Jones, B. and Peng, R.D. (2018). Projected trends in high‐mortality
heatwaves under different scenarios of climate, population, and adaptation in 82 US communities.
Climatic Change, 146, 455–470.

Anenberg, S.C., Horowitz, L.W., Tong, D.Q. and West J.J. (2010). An estimate of the global burden of
anthropogenic ozone and fine particulate matter on premature human Mortality using atmospheric
modeling. Env. Health Persp., 118, 1189–1195.

Anenberg, S.C., Henze, D.K., Tinney, V., Kinney, P.L., Raich, W., Fann, N., Malley, C.S., Roman, H.,
Lamsal, L., Duncan, B. and Martin, R.V. (2018). Estimates of the global burden of ambient PM 2.5,
ozone, and NO2 on asthma incidence and emergency room visits. Environmental Health Perspectives,
126(10), p.107004.

Anthony, K.W., von Deimling, T.S., Nitze, I., Frolking, S., Emond, A., Daanen, R., Anthony, P., Lindgren,
P., Jones, B. and Grosse, G. (2018). 21st-century modeled permafrost carbon emissions accelerated
by abrupt thaw beneath lakes. Nature Communications, 9, 1-11.

Archibald, A., O’Connor, F., Archer-Nicholls, S., Chipperfield, M., Dalvi, M., Folberth, G., Dennison,
F., Dhomse, S., Griffiths, P., Hardacre, C., Hewitt, A., Hill, R., Johnson, C., Keeble, J., Köhler, M.,
Morgenstern, O., Mulchay, J., Ordóñez, C., Pope, R., Rumbold, S., Russo, M., Savage, N., Sellar,
A., Stringer, M., Turnock, S., Wild, O. and Zeng, G. (2019). Description and evaluation of the UKCA
stratosphere-troposphere chemistry scheme (StratTrop vn 1.0) implemented in UKESM1. Geosci.
Model Dev. Discuss., doi:10.5194/gmd-2019-246.

Armstrong, B., Sera, F., Vicedo-Cabrera, A.M., Abrutzky, R., Åström, D.O., Bell, M.L.,

Chen, B-Y., de Sousa Zanotti Stagliorio Coelho, M., Correa, P.M., Dang, T.N., Diaz, M.H., Dung,
Global Methane Assessment / REFERENCES 129

D.V., Forsberg, B., Goodman, P., Guo, Y-L.L., Guo, Y., Hashizume, M., Honda, Y., Indermitte, E.,
Íñiguez, C., Kan, H., Kim, H., Kyselý, J., Lavigne, E., Michelozzi, P., Orru, H., Ortega, N.V., Pascal, M.,
Ragettli, M.S., Saldiva, P.H.N., Schwartz, J., Scortichini, M., Seposo, X., Tobias, A., Tong, S., Urban,
A., De la Cruz Valencia, C., Zanobetti, A., Zeka, A. and Gasparrini, A. (2019). The role of humidity in
associations of high temperature with mortality: a multicountry, multicity study. Environmental Health
Perspectives, 127(9), 097007.

Aunan, K., Fang, J., Hu, T., Seip, H.M., and Vennemo, H. (2006). Climate change and air quality-
measures with co-benefits in China. Environ. Sci. Technol., 40, 4822-4829.

Avnery, S., Mauzerall, D.L., Liu, J. and Horowitz, L.W. (2011). Global crop yield reductions due to
surface ozone exposure: 1. Year 2000 crop production losses and economic damage, Atmospheric
Environment, 45, 2284-2296.

Bajželj, B., Richards, K.S., Allwood, J.M., Smith, P., Dennis, J.S., Curmi, E. and Gilligan, C.A. (2014).
Importance of food-demand management for climate mitigation. Nature Climate Change, 4, 924–929.

Baray, S., Darlington, A., Gordon, M., Hayden, K.L., Leithead, A., Li, S.M., Liu, P.S.K., Mittermeier,
R.L., Moussa, S.G., O’Brien, J., Staebler, R., Wolde, M., Worthy, D., McLaren, R. (2018). Quantification
of Methane Sources in the Athabasca Oil Sands Region of Alberta by Aircraft Mass Balance. Atmos.
Chem. Phys., 18, 7361– 7378.

Bauer, N., Calvin, K., Emmerling, J., Fricko, O., Fujimori, S., Hilaire, J., Eom, J., Krey, V., Kriegler, E.,
Mouratiadou, I. and de Boer, H.S. (2017). Shared socio-economic pathways of the energy sector–
quantifying the narratives. Global Environmental Change, 42, pp.316-330.

Bender, M., Sowers, T. and Brook, E. (1997). Ice core isotope data constrain methane emissions.
Proceedings of the National Academy of Sciences, 94 (16), 8343-8349.

Benmergui, J.S. (2019). Innovations in Modeling and Statistical Analysis for Greenhouse Gas Flux
Monitoring. Doctoral dissertation, Harvard University, Graduate School of Arts & Sciences.

Bentayeb, M., Wagner, V., Stempfelet, M., Zins, M., Goldberg, M., Pascal, M., Larrieu, S., Beaudeau,
P., Cassadou, S., Eilstein, D., Filleul, L., Le Tertre, A., Medinaa, S., Pascal, L., Prouvost, H., Quénel,
P., Zeghnoun, A. and Lefranc, A. (2015). Association between long-term exposure to air pollution and
mortality in France: a 25-year follow-up study. Environ Int., 85, 5–14.

Bergamaschi, P., Frankenberg, C., Meirink, J.F., Krol, M., Dentener, F., Wagner, T., Platt, U., Kaplan,
J.O., Körner, S., Heimann, M., Dlugokencky, E.J. and Goede, A. (2007). Satellite chartography of
atmospheric CH4 from SCIAMACHY on board ENVISAT: 2. Evaluation based on inverse model
simulations. J. Geophys. Res.-Atmos., 112, D02304.

Bergamaschi, P., Frankenberg, C., Meirink, J.F., Krol, M., Villani, M.G., Houweling, S., Dentener, F.,
Dlugokencky, E.J., Miller, J.B., Gatti, L.V., Engel, A., and Levin, I. (2009). Inverse modeling of global
and regional CH4 emissions using SCIAMACHY satellite retrievals. J. Geophys. Res., 114, D22301.

Bergamaschi, P., Houweling, S., Segers, A., Krol, M., Frankenberg, C., Scheepmaker, R.A.,
Dlugokencky, E., Wofsy, S.C., Kort, E.A., Sweeney, C., Schuck, T., Brenninkmeijer, C., Chen, H., Beck,
V. and Gerbig, C. (2013). Atmospheric CH4 in the first decade of the 21st century: Inverse modeling
analysis using SCIAMACHY satellite retrievals and NOAA surface measurements. J. Geophys. Res.-
Atmos., 118, 7350–7369.

Bignell, D.E., Eggleton, P., Nunes, L., Thomas, K.L. (2017). Termites as mediators of carbon fluxes
in tropical forest: Budgets for carbon dioxide and methane emissions. In Watt, A.D., Stork, N.E. and
Hunter, M.D. (eds.). Forests and Insects. Chapman & Hall, London, UK.

Blake, D.R., Mayer, E.W., Tyler, S.C., Makide, Y., Montague, D.C. and Rowland, F.S. (1982). Global
increase in atmospheric methane concentrations between 1978 and 1980. Geophys. Res. Lett., 9:
477-480.

Bock, M., Schmitt, J., Beck, J., Seth, B., Chappellaz, J. and Fischer, H. (2017). Glacial/interglacial
wetland, biomass burning, and geologic methane emissions constrained by dual stable isotopic
Global Methane Assessment / REFERENCES 130

CH4 ice core records. Proceedings of the National Academy of Sciences, 114 (29) E5778-E5786.

Borgonovo, F., Conti, C., Lovarelli, D., Ferrante, V. and Guarino, M. (2019). Improving the sustainability
of dairy slurry by a commercial additive treatment. Sustainability, 11, 4998.

Brandt, A.R., Heath, G.A., Kort, E.A., O’Sullivan, F., Pétron, G., Jordaan, S.M., Tans, P., Wilcox, J.,
Gopstein, A.M., Arent, D., Wofsy, S., Brown, N.J., Bradley, R., Stucky, G.D., Eardley, D. and Harriss,
R. (2014). Methane Leaks from North American Natural Gas Systems, Science, 343, 733–735.

Bröde, P., Fiala, D., Lemke, B. and Kjellstrom, T. (2017). Estimated work ability in warm outdoor
environments depends on the chosen heat stress assessment metric. Int J Biometeorol. https://doi.
org/10.1007/s00484-017-1346-9 (accessed 6 March 2021).

Brune, A. (2010). Methanogenesis in the digestive tracts of insects. In Timmis, K.N. (ed.). Handbook
of Hydrocarbon and Lipid Microbiology, Vol 8. Springer Verlag, Berlin, Germany.

Bryngelsson, D., Wirsenius, S., Hedenus, F. and Sonesson, U. (2016). How can the EU climate targets
be met? A combined analysis of technological and demand-side changes in food and agriculture.
Food Policy, 59, 152–164.

Buchwitz, M., Schneising, O., Reuter, M., Heymann, J., Krautwurst, S., Bovensmann, H., Burrows,
J.P., Boesch, H., Parker, R.J., Somkuti, P. and Detmers, R.G. (2017). Satellite-derived methane hotspot
emission estimates using a fast data-driven method. Atmospheric Chemistry and Physics, 17, 5751.

Buzan, J.R. and Huber, M. (2020). Moist heat stress on a hotter Earth. Annual Review of Earth and
Planetary Sciences, 48.

Calvin, K.V., Patel, P.L., Clarke, L.E., Asrar, G.R., Bond-Lamberty, B., Cui, Y., Di Vittorio, A., Dorheim,
K.R., Edmonds, J.A., (2019). GCAM v5.1: Representing the linkages between energy, water, land,
climate, and economic systems. Geoscientific Model Development, 12, 1–22.

Cambaliza, M.O.L., Bogner, J.E., Green, R.B., Shepson, P.B., Harvey, T.A., Spokas. K.A., Stirm, B.H.,
Corcoran, M., Hartin, C., Hejazi, M., Horowitz, R., Iyer. G., Kyle, P., Kim, S., Link, R., McJeon, H., Smith,
S.J., Snyder, A., Waldhoff, S. and Wise, M. (2017). Field measurements and modeling to resolve m2 to
km2 CH4 emissions for a complex urban source: An Indiana landfill study, Elementa: Science of the
Anthropocene, 5(36).

Camilleri, A.R., Larrick, R.P., Hossain, S. and Patino-Echeverri, D. (2017). Consumers underestimate
the emissions associated with food but are aided by labels. Nat. Clim. Change, 9, 53–8.

Carey, I.M., Atkinson, R.W., Kent, A.J., van Staa, T., Cook, D.G. and Anderson, H.R. (2013). Mortality
Associations with Long-Term Exposure to Outdoor Air Pollution in a National English Cohort. Am J
Respir Crit Care Med., 187, 1226–123.

Carleton, T.A., Jina, A., Delgado, M.T., Greenstone, M., Houser, T., Hsiang, S.M., Hultgren, A., Kopp,
R.E., McCusker, K.E., Nath, I.B., Rising, J., Rode, A., Seo, H.K., Viaene, A., Yuan, J, and Zhang, A.T.
(2020). Valuing the Global Mortality Consequences of Climate Change Accounting for Adaptation Costs
and Benefits. NBER Working Paper No. 27599. National Bureau of Economc Research, Cambridge,
MA, US. http://www.nber.org/papers/w27599 (accessed 6 March 2020)

Carmichael, M.J., Bernhardt, E.S., Bräuer, S.L. and Smith, W.K. (2014). The role of vegetation in
methane flux to the atmosphere: should vegetation be included as a distinct category in the global
methane budget? Biogeochemistry, 119(1-3), 1-24.

Center for International Earth Science Information Network, Columbia University, Food and Agriculture
Programme of the United Nations, and Centro Internacional de Agricultura Tropical (2016). Gridded
Population of the World, Version 4 (GPWv4): Population Count Grid. NASA Socioeconomic Data and
Applications Center (SEDAC), Palisades, NY, US.

Chae, Y. and J. Park. (2011). Quantifying costs and benefits of integrated environmental strategies of
air quality management and greenhouse gas reduction in the Seoul Metropolitan Area. Energy Policy,
39, 5296-5308.
Global Methane Assessment / REFERENCES 131

Challinor, A.J., Watson, J., Lobell, D.B., Howden, S.M., Smith, D.R. and Chhetri, N. (2014). A meta‐
analysis of crop yield under climate change and adaptation. Nature Climate Change, 4, 287–291.

Chappellaz J., Raynaud D., Blunier T. and Stauffer B. (2000) The Ice Core Record of Atmospheric
Methane. In Khalil M.A.K. (ed.). Atmospheric Methane. Springer Verlag, Berlin, Germany.

Chappellaz, J., Barnola, J., Raynaud, D., Korotkevich, Y.S. and Lorius, C. (1990). Ice-core record of
atmospheric methane over the past 160,000 years. Nature, 345, 127–131.

Chan, E., Worthy, D.E., Chan, D., Ishizawa, M., Moran, M.D., Delcloo, A. and Vogel, F. (2020). Eight-Year
Estimates of Methane Emissions from Oil and Gas Operations in Western Canada are Nearly Twice
those Reported in Inventories. Environmental Science & Technology, 54, 23, 14899–14909.

Cifuentes, L., Borja-Aburto, V.H., Gouveia, N., Thurston, G. and Davis, D.L. (2001). Hidden health
benefits of greenhouse gas mitigation. Science, 293.

Clark, M.A., Springmann, M., Hill, J. and Tilman, D. (2019). Multiple health and environmental impacts
of foods, Proceedings of the National Academy of Sciences, 116(46), 23357–23362.

Collins, W.J., Sitch, S. and Boucher, O. (2010). How vegetation impacts affect climate metrics for ozone
precursors. J. Geophys. Res., 115, D23308.

Collins, M., Knutti, R., Arblaster, J., Dufresne, J.-L., Fichefet, T., Friedlingstein, P., Gao, X., Gutowski,
W.J., Johns, T., Krinner, G., Shongwe, M., Tebaldi, C., Weaver, A.J. and Wehner, M. (2013). Long-term
Climate Change: Projections, Commitments and Irreversibility. In Stocker, T.F., Qin, D.. Plattner, G.-
K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V. and Midgley, P.M. (eds.). Climate
Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment
Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge,
UK.

Collins, W.J., Fry, M.M., Yu, H., Fuglestvedt, J.S., Shindell, D.T. and West, J.J. (2013). Global and regional
temperature-change potentials for near-term climate forcers. Atmos. Chem. Phys., 13, 2471–2485.

Collins, W.J., Webber, C.P., Cox, P.M., Huntingford, C., Lowe, J., Sitch, S., Chadburn, S.E., Comyn-
Platt, E., Harper, A.B., Hayman, G. and Powell, T. (2018). Increased importance of methane reduction
for a 1.5 degree target. Environmental Research Letters, 13, 054003.

Conley, A.J., Westervelt, D.M., Lamarque, J.-F., Fiore, A.M., Shindell, D., Correa, G., Faluvegi, G. and
Horowitz, L.W. (2018). Multimodel surface temperature responses to removal of U.S. sulfur dioxide
emissions. J. Geophys. Res., 123, 2773–2796.

Conley, S., Faloona, I., Mehrotra, S., Suard, M., Lenschow, D.H., Sweeney, C., Herndon, S., Schwietzke,
S., Pétron, G., Pifer, J. and Kort, E.A. (2017). Application of Gauss’s theorem to quantify localized
surface emissions from airborne measurements of wind and trace gases. Atmospheric Measurement
Techniques, 10(9).

Cooper, M.D., Estop-Aragonés, C., Fisher, J.P., Thierry, A., Garnett, M.H., Charman, D.J., Murton, J.B.,
Phoenix, G.K., Treharne, R., Kokelj, S.V. and Wolfe, S.A. (2017). Limited contribution of permafrost
carbon to methane release from thawing peatlands. Nature Climate Change, 7, 507-511.

Crouse, D.L., Peters, P.A., Hystad, P., Brook, J.R., van Donkelaar, A., Martin, R.V.,

Villeneuve, P.J., Jerrett, M., Goldberg, M.S., Arden Pope III, C., Brauer, M., Robert D. Brook, R.D.,
Robichaud, A., Richard Menard, R. and Burnett, R.T. (2015). Ambient PM2.5, O3, and NO2 exposures
and associations with mortality over 16 years of follow-up in the Canadian Census Health and
Environment Cohort (CanCHEC). Environ. Health Persp., 123, 1180–1186.

Danabasoglu, G., Lamarque, J.-F., Bacmeister, J., Bailey, D.A., DuVivier, A.K., Edwards, J., Emmons,
L., Fasullo, J., Garcia, R., Gettelman, A., Hannay, C., Holland, M.M., Large, W.G., Lawrence, D.,
Lenaerts, J., Lindsay, K., Lipscomb, W., Lofverstrom, M., Mills, M., Neale, R., Oleson, K., Otto-
Bleisner, B., Phillips, A., Sacks, W., Tilmes, S., Vertenstein, M., Bertini, A., Deser, C., Fox-Kemper,
B., Kay, J.E., Kushner, P., Long, M.C., Mickelson, S., Moore, J.K., Nienhouse, E., Polvani, L., Rasch,
Global Methane Assessment / REFERENCES 132

P.J., and Strand, W.G. (2020). The Community Earth System Model version 2 (CESM2).  Accepted for
publication in JAMES.

Dean, J.F., Middelburg, J.J., Röckmann, T., Aerts, R., Blauw, L.G., Egger, M., Jetten, M.S., de Jong,
A.E., Meisel, O.H., Rasigraf, O. and Slomp, C.P. (2018). Methane feedbacks to the global climate
system in a warmer world. Reviews of Geophysics, 56, 207-250.

Di, Q., Wang, Y., Zanobetti, A., Wang, Y., Koutrakis, P., Choirat C., Dominici, F. and Schwartz, J.D.
(2017). Air pollution and mortality in the Medicare population, N Engl J Med., 376, 2513–2522.

Dlugokencky E. and Houweling S. (2015). Chemistry of the Atmosphere: Methane, In Encyclopedia of

Atmospheric Sciences, 2nd. Edition. Academic Press, Cambridge, MA, US.

Drew, J., Cleghorn, C., Macmillan, A. and Mizdrak, A. (2020). Healthy and Climate-Friendly Eating
Patterns in the New Zealand Context. Environmental Health Perspectives, 128,1, 17007.

Dunne, J.P., Horowitz, L.W., Adcroft, A.J., Ginoux, P., Held, I.M., John, J.G., Krasting, J.P., Malyshev,
S., Naik, V., Paulot, F., Shevliakova, E., Stock, C. A., Zadeh, N., Balaji, V., Blanton, C., Dunne, K.A.,
Dupuis, C., Durachta, J., Dussin, R., Gauthier, P.P.G., Griffies, S.M., Guo, H., Hallberg, R.W., Harrison,
M., He, J., Hurlin, W., McHugh, C., Menzel, R., Milly, P.C.D., Nikonov, S., Paynter, D. J., Ploshay,
J., Radhakrishnan, A., Rand, K., Reichl, B.G., Robinson, T., Schwarzkopf, D.M., Sentman, L.T.,
Underwood, S., Vahlenkamp, H., Winton, M., Wittenberg, A.T., Wyman, B., Zeng, Y. and Zhao, M.
(2020). The GFDL Earth System Model version 4.1 (GFDL-ESM4.1): Model description and simulation
characteristics. J. Adv. Modeling Earth Sys. https://doi.org/10.1029/2019MS002015 (accessed 6
March 2021).

Duren, R.M., Thorpe, A.K., Foster, K.T., Rafiq, T., Hopkins, F.M., Yadav, V., Bue, B.D..Thimpson, D.R.,
Conley, S., Colombi, N.K., Frankenberg, C., McCubbin, I.B., Eastwood, M.L. Falk, M., Herner, J.D.,
Croes, B.E., Green, R.D. and Miller, C.E. (2019). California’s methane super-emitters. Nature, 575,
180-185.

Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K. Adler, A., Baum,
I., Brunner, S., Eikemeier, P., Kriemann, B., Savolainen, J., Schlömer, S., von Stechow, C. and Zwickel,
T. (2014). Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III
to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge, UK.

Etheridge, D.M., Steele, L.P., Francey, R.J. and Langenfeld, R.L. (1998). Atmospheric methane
between 1000 A.D. and present: Evidence of anthropogenic emissions and climatic variability. J.
Geophys. Res., 103, D13, 15,979-15,993.

Etiope, G. and Schwietzke, S. (2019). Global geological methane emissions: an update of top-down
and bottom-up estimates. Elem. Sci. Anth., 7, 47. 

Etminan, M., Myhre, G., Highwood, E.J. and Shine, K.P. (2016). Radiative forcing of carbon dioxide,
methane, and nitrous oxide: A significant revision of the methane radiative forcing. Geophysical
Research Letters, 43, 12-614.

EUROSTAT. (2016). Hospital discharges by diagnosis. Publications Office of the European Union,
Luxembourg.

Fann, N., Coffman, E., Timin, B. and Kelly, J.T. (2018). The estimated change in the level and
distribution of PM2.5-attributable health impacts in the United States: 2005-2014. Environ Res., 167,
506-514.

Fann, N., Baker, K.R., Chan, E.A.W., Eyth, A., Macpherson, A., Miller, E. and Snyder, J. (2018)
Assessing Human Health PM 2.5 and Ozone Impacts from U.S. Oil and Natural Gas Sector Emissions
in 2025. Environ. Sci. Technol., 52, 8095–8103.

Food and Agriculture Organization (2014). Food wastage footprint: Full cost accounting. Food and
Agriculture Organization of the United Nations, Rome, Italy.
Global Methane Assessment / REFERENCES 133

Food and Agriculture Organization of the United Nations (2017). Global Livestock Environmental
Assessment Model (GLEAM). Food and Agriculture Organization of the United Nations, Rome, Italy.
www.fao.org/gleam/en/ (accessed 6 March 2021).

Food and Agriculture Organization of the United Nations (2019). State of Food and Agriculture
(SOFA), 2019. Food and Agriculture Organization of the United Nations, Rome, Italy. http://www.fao.
org/state-of-food-agriculture/2019/en/ (accessed 6 March 2021)

Food and Agriculture Organization of the United Nations, International Fund for Agricultural
Development, United Nations Children’s Fund, World Food Programme and World Health Organization
(2017). The State of Food Security and Nutrition in the World: Building Resilience for Peace and Food
Security. Food and Agriculture Organization of the United Nations, Rome, Italy.

Food and Agriculture Organization of the United Nations, International Fund for Agricultural
Development, United Nations Children’s Fund, World Food Programme and World Health Organization
(2020). The State of Food Security and Nutrition in the World 2020. Transforming food systems for
affordable healthy diets. Food and Agriculture Organization of the United Nations, Rome, Italy. https://
doi.org/10.4060/ca9692en (accessed 6 March 2021).

Field, C.B., Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D., Bilir, T.E.

Chatterjee, M., Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N.. MacCracken,
S., Mastrandrea, P.R. and White, L.L. (eds.). (2014). Climate Change 2014: Impacts, Adaptation,
and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK.

Fiore, A.M., Jacob, D.J., Field, B.D., Streets, D.G., Fernandes, S.D. and Jang, C. (2002). Linking
ozone pollution and climate change: The case for controlling methane. Geophysical Research Letters,
29, 25-1.

Fiore, A.M., West, J.J., Horowitz, L.W., Naik, V. and Schwarzkopf, M.D. (2008). Characterizing the
tropospheric ozone response to methane emission controls and the benefits to climate and air quality.
J. Geophys. Res., 113, 1–16.

Frankenberg, C., Meiring, J.F., Bergamaschi, P., Goede, A.P.H., Heimann, M., Korner, S., Platt, U., van
Weele, M., and Wagner, T. (2006). Satellite chartography of atmospheric methane from SCIAMACHY on
board ENVISAT: Analysis of the years 2003 and 2004. JGR: Atmospheres. Doi:10.1029/2005JD006235.

Frankenberg, C., Thorpe, A.K., Thompson, D.R., Hulley, G., Kort, E.A., Vance, N., Borchardt, J.,
Krings, T., Gerilowski, K., Sweeney, C., Conley, S., Bue, B.D., Aubrey, A.D., Hook, S. and Green, R.O.
(2016). Airborne methane remote measurements reveal heavy-tail flux

distribution in Four Corners region. Proc. Natl Acad. Sci., 113, 9734–9739.

Fu, B., Gasser, T., Li, B. and Tao, S. (2020) Short-lived climate forcers have long-term climate impacts
via the carbon–climate feedback. Nat. Clim. Chang., 10, 851–855.

Fujimori, S., Hasegawa, T., Masui, T., Takahashi, K., Herran, D. S., Dai, H., Hijioka, Y. and Kainuma,
M. (2017). SSP3: AIM implementation of Shared Socioeconomic Pathways. Global Environmental
Change, 42, 268–283.

FUSIONS. (2016). Estimates of European food waste levels. IVL Swedish Environmental Research
Institute, Stockholm, Sweden. http://www.eu-fusions.org/phocadownload/Publications/
Estimates%20of%20European%20food%20waste%20levels.pdf (accessed 6 March 2021).

Gao, J., Kovats, S., Vardoulakis, S., Wilkinson, P., Woodward, A., Li, J., Gu, S., Liu, X., Wu, H.,
Wang, J., Song, X., Zhai, Y., Zhao, J. and Liu, Q. (2018). Public health co-benefits of greenhouse gas
emissions reduction: A systematic review. Science of the Total Environment, 627, 388-402.

Garrone, P., Melacini, M. and Perego, A. (2014). Opening the black box of food waste reduction. Food
Pol., 46, 129-139.
Global Methane Assessment / REFERENCES 134

Gasbarra, D., Toscano, P., Famulari, D., Finardi, S., Di Tommasi, P., Zaldei, A., Carlucci, P., Magliulo,
E. and Gioli, B. (2019). Locating and quantifying multiple landfills methane emissions using aircraft
data. Environmental Pollution, 254, Part B, 112987, ISSN 0269-7491.

Gasser, T., Peters, G.P., Fuglestvedt, J.S., Collins, W.J., Shindell, D.T. and Ciais, P. (2017). Accounting
for the climate-carbon feedback in emission metrics. Earth System Dynamics, 8(2), 235-253.

Gasparrini, A., Guo, Y.M., Hashizume, M., Lavigne, E., Zanobetti, A., Schwartz, J., Tobias, A., Tong,
S.L., Rocklov, J., Forsberg, B., Leone, M., De Sario, M., Bell, M.L., Guo, Y.L.L., Wu, C.F., Kan, H.,
Yi, S.M., Coelho, M., Saldiva, P.H.N., Honda, Y., Kim, H. and Armstrong, B. (2015). Mortality risk
attributable to high and low ambient temperature: a multicountry observational study. The Lancet,
386, 369–375.

Gedney, N., Cox, P.M. and Huntingford, C. (2004). Climate feedback from wetland methane emissions.
Geophysical Research Letters, 31(20).

Geoffroy, O., Saint-Martin, D., Bellon, G., Voldoire, A., Olivié, D.J.L. and Tytéca, S. (2013). Transient
climate response in a two-layer energy-balance model. Part II: Representation of the efficacy of
deep-ocean heat uptake and validation for CMIP5 AOGCMs. J. Climate., 26, 1859-1876.

Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A. and Tempio, G.
(2013). Tackling climate change through livestock – A global assessment of emissions and mitigation
opportunities. Food and Agriculture Organization of the United Nations, Rome, Italy.

Gettelman, A., Mills, M.J., Kinnison, D.E., Garcia, R.R., Smith, A.K., Marsh, D.R., Tilmes, S., Vitt, F.,
Bardeen, C.G., McInerny, J., Liu, H.-L., Solomon, S.C., Polvani, L.M., Emmons, L.K., Lamarque, J.-
F., Richter, J.H., Glanville, A.S., Bacmeister, J.T., Phillips, A.S., Neale, R.B., Simpson, I.R., DuVivier,
A.K., Hodzic, A. and Randel, W.J. (2019). The Whole Atmosphere Community Climate Model Version
6 (WACCM6). J. Geophys. Res. https://doi.org/10.1029/2019JD030943 (accessed 6 March 2021).

Gillett, N.P., Kirchmeier-Young, M., Ribes, A., Shiogama, H., Hegerl, G.C., Knutti, R., Gastineau, G.,

John, J.G., Li, L., Larissa Nazarenko, L., Rosenbloom, N.,  Seland, Ø., Wu, T., Yukimoto, S. and Ziehn,
T. 2021. Constraining human contributions to observed warming since the pre-industrial period.
Nature Climate Change. https://doi.org/10.1038/s41558-020-00965-9 (accessed 19 Feb. 2021).

Gilligan, J., Dietz, T., Gardner, G., Stern, P. and Vandenbergh, M. (2010). The behavioural wedge.
Significance, 7: 17–20.

Global Burden of Disease Collaborative Network. (2017). Global Burden of Disease Study 2017 (GBD
2017) Results. The Institute for Health Metrics and Evaluation, Seattle, WA, US.

Gorchov Negron, A.M., Kort, E.A., Conley, S.A. and Smith, M.L. (2020). Airborne Assessment of
Methane Emissions from Offshore Platforms in the US Gulf of Mexico. Environmental Science &
Technology, 54, 5112–5120.

Graff Zivin, J. and Neidell, M. (2014). Temperature and the Allocation of Time: Implications for Climate
Change. J. Labor Econ., 32(1), 1–26.

Green, R., Milner, J., Dangour, A.D. and Haines, A. (2015). The potential to reduce greenhouse gas
emissions in the UK through healthy and realistic dietary change. Climatic Change, 129, 253.

Gustavsson, J., Cederberg, C., Sonesson, U., van Otterdijk, R. and Meybeck, A. (2011). Global Food
Losses and Food Waste – Extent, Causes and Prevention. Food and Agriculture Organization of the
United Nations, Rome, Italy.

Gvakharia, A., Kort, E.A., Brandt, A., Peischl, J., Ryerson, T.B., Schwarz, J.P., Smith. M.L. and
Sweeney, C. (2017). Methane, black carbon, and ethane emissions from natural gas flares in the
Bakken Shale, North Dakota. Environmental Science & Technology, 51(9), 5317–5325.

Haines, A. (2012). Health benefits of a low carbon economy. Public Health, 126 (Suppl. 1), pp. S33-9.

Haines, A., Amann, M., Borgford-Parnell, N., Leonard, S., Kuylenstierna, J.C.I. and Shindell, D. (2017).
Short-lived climate pollutant mitigation and the sustainable development goals. Nature Climate
Global Methane Assessment / REFERENCES 135

Change, 7, 863–869.

Hammitt, J.K. and Robinson, L.A. (2011). The income elasticity of the value per statistical life:
transferring estimates between high and low income populations. Journal of Benefit-Cost Analysis,
2(1), 1–29.

Hansen, J., Sato, M., Ruedy, R., Nazarenko, L., Lacis, A., Schmidt, G.A., Russell, G., Aleinov, I., Bauer,
M., Bauer, S.E., Bell, N., Cairns, B., Canuto, V., Chandler, M., Cheng, Y., Del Genio, N., Faluvegi, G.,
Fleming, E., Friend, A., Hall, T., Jackman, H., Kelley, M., Kiang, N., Koch, D., Lean, J., Lo, K., Menn,
S., Miller, R.L., Minnis, P., Novakov, T., Oinas, V., Perlwitz, Ja., Rind, D., Romanou, A., Shindell, D.,
Stone, P., Sun, S., Tausnev, N., Thresher, D., Wielieki, D., Wong, T., Yao, M. and Zhang, S. (2005).
Efficacy of climate forcings. J. Geophys. Res., 110, D18104.

Harmsen, M.J.H.M., van Den Berg, M., Krey, V., Luderer, G., Marcucci, A., Strefler, J. and Van Vuuren,
D.P. (2016). How climate metrics affect global mitigation strategies and costs: a multi-model study.
Climatic Change, 136, 203–216.

Harmsen, M.J.H.M., van Vuuren, D.P., Nayak, D.R., Hof, A.F., Höglund-Isaksson, L., Lucas, P.L.,
Nielsen, J.B., Smith, P. and Stehfest, E., (2019). Long-term marginal abatement cost curves of non-
CO2 greenhouse gases. Environmental Science & Policy, 99, 136–149.

Harmsen, M.J.H.M., van Vuuren, D.P., Bodirsky B.L., Chateau, J., Durand-Lasserve, O., Drouet, L.,
Fricko, O., Fujimori, S., Gernaat, D.E.H.J., Hanaoka, T., Hilaire, J., Keramidas, K., Luderer, G., Moura,
M.C.P., Sano, F., Smith, S.J. and Wada, K. (2019b). The role of methane in future climate strategies:
mitigation potentials and climate impacts. Climatic Change, 163:1409–1425.

https://doi.org/10.1007/s10584-019-02437-2 (accessed 6 March 2021).

Hedenus, F., Wirsenius, S. and Johansson, D.J.A. (2014). The importance of reduced meat and dairy
consumption for meeting stringent climate change targets. Climatic Change, 124(1–2): 79–91.

Héroux, M.E., Anderson, H.R., Atkinson, R., Brunekreef, B., Cohen, A., Forastiere, F., Hurley, F.,
Katsouyanni, K., Krewski, D., Krzyzanowski, M. and Künzli, N. (2015). Quantifying the health impacts
of ambient air pollutants: recommendations of a WHO/Europe project. Int J Public Health, 60, 619–27.

Hmiel, B., Petrenko, V.V., Dyonisius, M.N., Buizert, C., Smith, A.M., Place, P.F., Harth, C., Beaudette,
R., Hua, Q., Yang, B., Vimont, I., Michel, S.E., Severinghaus, J.P., Etheridge, D., Bromley, T.,
Schmitt, J., Faïn, X., Weiss, R.F. and Dlugokencky, E. (2020). Preindustrial 14 CH4 indicates greater
anthropogenic fossil CH4 emissions. Nature, 578, 409–412.

Höglund-Isaksson, L., Gómez-Sanabria, A. Klimont, Z., Rafaj, P. and Schöpp, W. (2020). Technical
potentials and costs for reducing global anthropogenic methane emissions in the 2050 timeframe
–results from the GAINS model. Environ. Res. Comm., 2, 025004.

Holtsmark, B. (2012). Harvesting in boreal forests and the biofuel carbon debt. Climatic Change, 112,
415–428.

Honda, Y., Kondo, M., McGregor, G., Kim, H., Guo, Y.L., Hijioka, Y., Yoshikawa, M., Oka, K., Takano,
S., Hales, S. and Kovats, R. (2014). Heat-related mortality risk model for climate change impact
projection. Environmental Health and Preventive Medicine, 19, 56–63.

Horowitz, L.W., Naik, V., Paulot, F., Ginoux, P.A., Dunne, J.P., Mao, J., Schnell, J., Chen, X., He, J.,
John, J.G. and Lin, M. (2020). The GFDL global atmospheric chemistry‐climate model AM4. 1: Model
description and simulation characteristics. Journal of Advances in Modeling Earth Systems, 12(10),
e2019MS00203.

Hristov, A.N., Oh, J., Giallongo, F., Frederick, T.W., Harper, M.T., Weeks, H.L., Branco, A.F., Moate,
P.J., Deighton, M.H., Williams, S.R.O. and Kindermann, M. (2015). An inhibitor persistently decreased
enteric methane emission from dairy cows with no negative effect on milk production. Proceedings
of the National Academy of Sciences, 112, 10663–10668.

Hsiang, S., Kopp, R., Jina, A., Hsiang, S., Kopp, R., Jina, A., Rising, A., Delgado, M., Mohan, S.,
Global Methane Assessment / REFERENCES 136

Rasmussen, D.J., Muir-Wood, R., Wilson, P., Oppenheimer, M., Larsen, K. and Houser, T. (2017).
Estimating economic damage from climate change in the United States. Science, 356(6345), 1362–
1369.

International Energy Agency (2020). Methane Tracker. International Energy Agency, Paris, France.
https://www.iea.org/reports/methane-tracker (accessed 6 March 2021).

International Labour Organization (2019). Working on a warmer planet: The impact of heat stress on
labour productivity and decent work. International Labour Office, Geneva, Switzerland.

International Labour Organization (2020). ILOSTAT explorer: Statutory nominal gross monthly minimum
wage – Harmonized series – Annual, 2020. International Labour Office, Geneva, Switzerland. https://
www.ilo.org/shinyapps/bulkexplorer0/?lang=en&segment=indicator&id=EAR_4MMN_CUR_NB_A
(accessed 6 March 2021).

International Futures (Ifs) modeling system, Version 7.58. Frederick S. Pardee Center for International
Futures, Josef Korbel School of International Studies, University of Denver, Denver, CO. http://www.
ifs.du.edu/ifs/frm_MainMenu.aspx (accessed 6 March 2021).

Intergovernmental Panel on Climate Change (2018). Summary for Policymakers. In Masson-

Delmotte, V., Masson Delmotte, V., Zhai, P., Portner, H.O., Roberts, D.C., Skea, J., Shukla, P.R.,
Pirani, A., Mouffouma-Okia, W., Pean, C., Pidcock, R., Connors, S., Matthews, J.B.R., Chen, Y.,
Zhou, X., Gomis, M.I., Lonnoy, E., Maycock, T., Tignor, M. and Waterfield, T. (eds.). Global warming of
1.5°C. An IPCC special report on the impacts of global warming of 1.5°C above pre-industrial levels
and related global greenhouse gas emission pathways, in the context of strengthening the global
response to the threat of climate change, sustainable development, and efforts to eradicate poverty.
World Meteorological Organisation, Geneva, Switzerland.

Ishangulyyev, R., Kim, S. and Lee, S.H. (2019). Understanding Food Loss and Waste-Why Are We
Losing and Wasting Food? Foods, 8, 297.

Jackson, R.B., Saunois, M., Bousquet, P., Canadell, J.G., Poulter, B., Stavert, A.R., Bergamaschi,
P., Niwa, Y., Segers, A. and Tsuruta, A. (2020). Increasing anthropogenic methane emissions arise
equally from agricultural and fossil fuel sources. Environ. Res. Lett., 15, 071002.

Jackson R.B., Solomon E.I., Canadell J.G., Cargnello M. and Field C.B. (2019). Comment, Methane
Removal and Atmospheric Restoration. Nature Sustainability, 2, 436–438.

Jackson R.B., Solomon E.I., Canadell J.G., Cargnello M., Field C.B. and Abernethy S. (2020). Reply
to: Practical constraints on atmospheric methane removal. Nature Sustainability, 3, 358–359.

Jacob, D., Turner, A., Maasakkers, J., Sheng, J-X., Sun, K., Liu, X., Chance, K., Aben, I., McKeever,
J. and Frankenberg, C. (2016). Satellite observations of atmospheric methane and their value for
quantifying methane emissions. Atmospheric Chemistry and Physics, 16. 14371–14396. 10.5194/
acp-16-14371-2016.

Jerrett, M., Burnett, R.T., Beckerman, B.S., Turner, M.C., Krewski, D., Thurston, G., Martin,R.V., van
Donkelaar, A., Hughes, E., Shi, Y., Gapstur, S.M., Thun, M.J and Arden Pope III, C. (2013). Spatial
analysis of air pollution and mortality in California. Am J Respir Crit Care Med., 188, 593–599.

Johnson, M.R., Tyner, D.R., Conley, S., Schwietzke, S. and Zavala-Araiza, D. (2017). Comparisons of
Airborne Measurements and Inventory Estimates of Methane Emissions in the Alberta Upstream Oil
and Gas Sector. Environ. Sci. Technol., 51, 13008–13017.

Karion, A., Sweeney, C., Pétron, G., Frost, G., Hardesty, R.M., Kofler, J., Miller, B.M., Newberger, T.,
Wolter, S., Banta, R., Brewer, A., Dlugokencky, E., Lang, P., Montzka, S.A., Schnell, R., Tans, P., Trainer,
M., Zamora, R. and Conley, S. (2013). Methane emissions estimate from airborne measurements over
a western United States natural gas field. Geophys. Res. Lett., 40, 4393–4397.

Karl, D.M., Beversdorf, L., Björkman, K.M., Church, M.J., Martinez, A. and Delong, E.F. (2008).
Aerobic production of methane in the sea. Nature Geosci., 1, 473–478.
Global Methane Assessment / REFERENCES 137

Katsouyanni, K., Samet, J.M., Anderson, H.R., Atkinson, R., Le, A.T., Medina, S., Samoli, E., Touloumi,
G., Burnett, R.T., Krewski, D. and Ramsay, T. (2009). Air pollution and health: a European and North
American approach (APHENA). Research report. Health Effects Institute, Boston, MA, US.

Kelley, M., Schmidt, G.A., Nazarenko, L.S., Bauer, S.E., Ruedy, R., Russell, G.L., Ackerman, A.S.,
Aleinov, I., Bauer, M., Bleck, R., Canuto, V., Cesana, G., Cheng, Y., Clune, T.L., Cook, B.I., Cruz,
C.A., Del Genio, A.D.,  Elsaesser, G.S., Faluvegi, G., Kiang, N.Y., Kim, D., Lacis, A.A., Leboissetier,
A., LeGrande, A.N., Lo, K.K., Marshall, J., Matthews, E.E., McDermid, S., Mezuman, K., Miller,
R.L., Murray, L.T., Oinas, V., Orbe,  C., García‐Pando, C.P., Perlwitz, J.P., Puma, M.J., Rind, D., 
Romanou, A., Shindell, D.T., Sun, S., Tausnev, N., Tsigaridis, K., Tselioudis, G., Weng, E., Wu, J. and
Yao, M.S. (2020). GISS-E2.1: Configurations and Climatology. J. Adv. Modeling Earth Sys., 12(8),
e2019MS002025.

Kelso, M. (2018). Pipeline Incidents Continue to Impact Residents. Fractracker Alliance, 7. https://
www.fractracker.org/2018/12/pipeline-incidents-impact-residents/ (accessed 9 March 2021).

Keppler, F., Hamilton, J.T., Brass, M. and Röckmann, T. (2006). Methane emissions from terrestrial
plants under aerobic conditions. Nature, 439,187–91.

Kinley, R.D., de Nys, R., Vucko, M.J., Machado, L. and Tomkins, N.W. (2016). The red macroalgae
Asparagopsis taxiformis is a potent natural antimethanogenic that reduces methane production
during in vitro fermentation with rumen fluid. Animal Production Science, 56, 282–289.

Kirschke, S., Bousquet, P., Ciais, P., Saunois, S., Canadell, J.G., Dlugokencky, E.J., Bergamaschi,
P., Bergmann, D., Blake, D.R., Bruhwiler, L., Cameron-Smith, P., Castaldi, S., Chevallier, F., Feng,
L., Fraser, A., Heimann, M., Hodson, E.L., Houweling, S., Josse, B., Fraser, P.J., Krummel, P.B.,
Lamarque, J-F., Langenfelds, R.L., Le Quéré, C., Naik, V., O’Doherty, S., Palmer, P.I., Pison, I.,
Plummer, D., Poulter, B., Prinn, R.G., Rigby, M., Ringeval, B., Santini, M., Schmidt, M., Shindell,
D.T., Simpson, I.J., Spahni, R., Steele. L.P., Strode, S.A., Sudo, K., Szopa, S., van der Werf, G.R.,
Voulgarakis, A., van Weele, M., Weiss, R.F., Williams, J.E. and Zeng, G.(2013). Three decades of
global methane sources and sinks. Nature Geoscience, 6, 813–823.

Kirtman, B., Power, S.B., Adedoyin, J.A., Boer, G.J., Bojariu, R., Camilloni, I., Doblas-Reyes, F.J.,
Fiore, A.M., Kimoto, M., Meehl, G.A., Prather, M., Sarr, A., Schär, C., Sutton, R., van Oldenborgh,
G.J., Vecchi, G. and Wang, H.J. (2013). Near-term Climate Change: Projections and Predictability.
In Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y.,
Bex, V. and Midgley, P.M. (eds.). Climate Change 2013: The Physical Science Basis. Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.
Cambridge University Press, Cambridge, U.K.

Kjellstrom, T., Freyberg, C., Lemke, B., Otto, M. and Briggs, D. (2018) Estimating population heat
exposure and impacts on working people in conjunction with climate change. Int J Biometeorol, 62,
291–306.

Kjellstrom, T., Lemke, B., Otto, M., Hyatt, O. and Dear, K. (2014). Occupational Heat Stress. Contribution
to WHO project on Global assessment of the health impacts of climate change. Technical Report
2014:4, 1–52. Climate CHIP.

Knittel, N., Jury, M.W., Bednar-Friedl, B., Bachner, G. and Steiner, A.K. (2020). A global analysis of
heat-related labour productivity losses under climate change – implications for Germany’s foreign
trade. Climatic Change, 1–19.

Kort, E.A., Frankenberg, C., Costigan, K.R., Lindenmaier, R., Dubey, M.K. and Wunch, D. (2014). Four
corners: The largest US methane anomaly viewed from space. Geophys. Res. Lett., 41, 6898–6903.

Lackner, K.S. (2020). Practical Constraints on Atmospheric Methane Removal. Nature Sustainability,
3, 357.

Lavoie, T. N., Paul B. Shepson, Maria O. L. Cambaliza, Brian H. Stirm, Stephen Conley, Shobhit
Mehrotra, Ian C. Faloona, and David Lyon, Spatiotemporal Variability of Methane Emissions at Oil
and Natural Gas Operations in the Eagle Ford Basin. Environmental Science & Technology, 51 (14),
8001-8009, 2017.
Global Methane Assessment / REFERENCES 138

Lee, M., Nordio, F., Zanobetti, A., Kinney, P., Vautard, R. and Schwartz, J. (2014). Acclimatization
across space and time in the effects of temperature on mortality: a time-series analysis. Environmental
Health, 13, doi: 10.1186/1476-069x-13-89.

Lee, J.Y., Kim, H., Gasparrini, A., Armstrong, B., Bell, M.L., Lavigne, F.S.E., Abrutzky, R., Tong, S., de
Sousa Zanotti Stagliorio Coelho, M., Hilario, P., Matus Correal Nicolas Valdes Ortega, N.S.P., Kan,
H., Osorio Garcia., S., Kyselý, J., Urban,, A., Orru, H., Indermitte, E., Jaakkola, J.J.K., Ryti, N.R.I.,
Pascal, M., Goodman, P.G., Zeka, A., Michelozzi, P., Scortichini, M., Hashizume, M., Honda, Y.,
Hurtado, M., Cruz, J., Seposo., Nunes, B., Teixeira, J.P., Tobias, A., Íñiguez, C., Forsberg, B., Åström,
C., Vicedo-Cabrera, A.M., Ragettli, M.S., Guo, Y-L,L, Chen, B-Y., Zanobetti, A., Schwartz, J., Dang,
T.N., Van, D.D., Mayvaneh, F., Overcenco, A., Li, S. and Guo,, Y..(2019). Predicted temperature-
increase-induced global health burden and its regional variability. Env. Intl., 131, 105027.

Lenhart, K., Klintzsch, T., Langer, G., Nehrke, G., Bunge, M., Schnell, S. and Keppler, F. (2016).

Evidence for methane production by the marine algae Emiliania huxleyi. Biogeosciences, 13, 3163–
3174.

Li, D., Yuan, J. and Kopp, R.B. (2020). Escalating global exposure to compound heat-humidity
extremes with warming. Environmental Research Letters.

Li, X., Norman, H.C., Kinley, R.D., Laurence, M., Wilmot, M., Bender, H., de Nys, R. and Tomkins,
N. (2016). Asparagopsis taxiformis decreases enteric methane production from sheep. Animal
Production Science, 58, 681–688.

Luber, G. and McGeehin, M. (2008). Climate Change and Extreme Heat Events. American Journal of
Preventive Medicine, 35, 429–435.

Ma, W., Chen, R. and Kan, H. (2014). Temperature-related mortality in 17 large Chinese cities: How
heat and cold affect mortality in China. Environmental Research, 134, 127–133.

Maasakkers, J.D., Jacob, D., Sulprizio, M.P., Scarpelli, T.R., Nesser, H., Sheng, J.X., Zhang, Y.,
Hersher, M., Bloom, A.A., Bowman, K.W., Worden, J.R., Janssens-Maenhout, G. and Parker, R.J.
(2019). Global distribution of methane emissions, emission trends, and OH concentrations and
trends inferred from an inversion of GOSAT satellite data for 2010-2015. Atmos. Chem. Phys., 19,
7859–7881.

Machado, L., Magnusson, M., Paul, N.A., Kinley, R., de Nys, R. and Tomkins, N. (2016). Dose-
response effects of Asparagopsis taxiformis and Oedogonium sp. On in vitro fermentation and
methane production. J Appl Phycol., 28, 1443–1452.

Machado, L., Tomkins, N., Magnusson, M., Midgley, D.J., de Nys, R. and Rosewarne, C.P. (2018).
In Vitro Response of Rumen Microbiota to the Antimethanogenic Red Macroalga Asparagopsis
taxiformis, Microb. Ecol., 75, 811–818.

Malley, C.S., Henze, D.K., Kuylenstierna, J.C.I., Vallack, H.W., Davila, Y., Anenberg S.C., Turner, M.C.
and Ashmore, M.R. (2017). Updated global estimates of respiratory mortality in adults ≥ 30 years of
age attributable to long-term ozone exposure. Env. Health Persp., 125, 087021.

Marais, E.A., Jacob, D.J., Wecht, K., Lerot, C., Zhang, L., Yu, K., Kurosu, T.P., Chance, K., and
Sauvage, B. (2014). Anthropogenic emissions in Nigeria and implications for ozone air quality: a view
from space. Atmos. Environ., 99, 32–40.

Marten, A. and Newbold, S. (2012). Estimating the social cost of non-CO2 GHG emissions: Methane
and nitrous oxide. Energy Policy, 51, 957.

Mbow, C., Rosenzweig, C. ., Barioni, L.G., Benton, T.G., Herrero, M, Krishnapillai, M.,
Liwenga, E., Pradhan, P., Rivera-Ferre, M.G., Sapkota, T., Tubiello, F.N. and Xu, Y.
(2019). Food Security. In Shukla, P.R., Skea, J., Calvo Buendia, E., Masson-Delmotte, V.,
Pörtner, H.-O., Roberts, D.C., Zhai, P., Slade, R., Connors, S., van Diemen, R., Ferrat, M., Haughey,
E., Luz, S., Neogi, S., Pathak, M., Petzold, J., Portugal Pereira, J., Vyas, P., Huntley, E., Kissick,
K., Belkacemi, M. and Malley, J. (eds.). Climate Change and Land: an IPCC special report on
climate change, desertification, land degradation, sustainable land management, food security,
Global Methane Assessment / REFERENCES 139

and greenhouse gas fluxes in terrestrial ecosystems. Intergovernmental Panel on Climate Change,
Geneva, Switzerland.

Mehrotra, S., Faloona, I., Suard, M., Conley, S. and Fischer, M.L. (2017). Airborne methane emission
measurements for selected oil and gas facilities across California. Environmental Science &
Technology, 51, 12981–12987.

Meijer, Y. et al. (2019). Copernicus Mission Requirements Document, EOP-SM/3088/YM-ym, Issue 2.0.

Miller, S.M., Michalak, A.M., Detmers, R.G., Hasekamp, O.P., Bruhwiler, L.M.P., and Schwietzke,
S. (2019). China’s coal mine methane regulations have not curbed growing emissions. Nature
Communications, doi:10.1038/s41467-018-07891-7.

Mills, G., Sharps, K.,  Simpson, D., Pleijel, H., Frei, M., Burkey, K., Emberson, L., Uddling, J., Broberg,
M., Feng, Z., Kobayashi, K. and Agrawal, M. (2018). Closing the global ozone yield gap: Quantification
and cobenefits for multistress tolerance. Glob Change Biol., 24, 4869–4893.

Monteil, G., Houweling, S., Butz, A., Guerlet, S., Schepers, D., Hasekamp, O., Frankenberg, C.,
Scheepmaker, R., Aben, I. and Röckmann, T. (2013). Comparison of CH4 inversions based on 15
months of GOSAT and SCIAMACHY observations. J. Geophys. Res.-Atmos., 118, 11807–11823,
doi:10.1002/2013JD019760.

Myhre, G., Shindell, D., Bréon, F.-M., Collins, W., Fuglestvedt, J., Huang, J., Koch, D., Lamarque, J.-
F., Lee, ., Mendoza, B., Nakajima, T., Robock, A., Stephens, G., Takemura, T. and Zhang, H. (2013).
Anthropogenic and Natural Radiative Forcing. In Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M.,
Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V. Midgley, P.M. (eds.). Climate Change 2013:
The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK.

Naik, V., Voulgarakis, A., Fiore, A.M., Horowitz, L.W., Lamarque, J.-F., Lin, M., Prather, M.J., Young,
P.J., Bergmann, D., Cameron-Smith, P.J., Cionni, I., Collins, W.J., Dalsøren, S. B., Doherty, R., Eyring,
V., Faluvegi, G., Folberth, G.A., Josse, B., Lee, Y.H., MacKenzie, I. A., Nagashima, T., van Noije, T.P.C.,
Plummer, D.A., Righi, M., Rumbold, S.T., Skeie, R., Shindel, D.T., Stevenson, D.S., Strode, S., Sudo,
K., Szopa, S. and Zeng, G. (2013). Preindustrial to present-day changes in tropospheric hydroxyl
radical and methane lifetime from the Atmospheric Chemistry and Climate Model Intercomparison
Project (ACCMIP). Atmos. Chem. Phys., 13, 5277–5298.

National Academies of Sciences, Engineering and Medicine. (2018). Improving Characterization of

Anthropogenic Methane Emissions in the United States. National Academies Press, Washington, DC,
U.S. https://www.ncbi.nlm.nih.gov/books/NBK519293/ (accessed 22 March 2021).

Nauer, P.A., Hutley, L.B. and Arndt, S.K. (2018). Termite mounds mitigate half of termite methane
emissions. Proceedings of the National Academy of Sciences, 115, 13306–13311.

Nisbet, R.E., Fisher, R., Nimmo, R.H., Bendall, D.S., Crill, P.M., Gallego-Sala, A.V., Hornibrook, E.R.,
López-Juez, E., Lowry, D., Nisbet, P.B., Shuckburgh, E.F., Sriskantharajah,  S., Howe, C.J. and Nisbet,
E. G. (2009). Emission of methane from plants. Proceedings Biological Sciences, 276, 1347–1354.

Nisbet, E.G., Manning, M.R., Dlugokencky, E J., Fisher, R.E., Lowry, D., Michel, S.E., (2019). Very
strong atmospheric methane growth in the 4 Years 2014–2017: Implications for the Paris Agreement.
Global Biogeochemical Cycles, 33, 318–342.

Nisbet, E. G., Fisher, R. E., Lowry, D., France, J. L., Allen, G., Bakkaloglu, S., Lund Myhre, C., Platt,
S.M., Allen, G., Bousquet, P., Brownlow, R., Cain, M., France, J.L., Hermansen, O., Hossaini, R.,
Jones, A.E., Levin, I., Manning, A.C., Myhre, G., Pyle, J.A., Vaughn, B.H., Warwick, N.J. and White,
J.W.C. (2020). Methane mitigation: methods to reduce emissions, on the path to the Paris agreement.
Reviews of Geophysics, 58, e2019RG000675.

National Health and Family Planning Commission (2016), China health and family planning statistical
yearbook. Peking Union Medical College Publishing House, Beijing, China.

Organisation for Economic Co-operation and Development (2018). Cost-Benefit Analysis and the
Global Methane Assessment / REFERENCES 140

Environment:  Further Developments and Policy Use. Organisation for Economic Co-operation and
Development Publishing, Paris, France. https://doi.org/10.1787/9789264085169-en (accessed 9
March 2021).

Oh, Y., Zhuang, Q., Liu, L., Welp, L.R., Lau, M.C.Y., Onstott, T.C., Medvigy, D., Bruhwiler, L., Dlugokencky,
E.J., Hugelius, G., D’Imperio, L. and Elberling, B. (2020). Reduced net methane emissions due to
microbial methane oxidation in a warmer Arctic, Nat. Clim. Chang. 10, 317–321, 2020.

Orellano, P., Quaranta, N., Reynoso, J., Balbi, B. and Vasquez, J. (2017). Effect of outdoor air pollution
on asthma exacerbations in children and adults: Systematic review and multilevel meta-analysis.
PloS One, 12, e0174050.

Pachauri, R.K., Allen, M.R., Barros, V.R., Broome, J., Cramer, W., Christ, R., Church, J.A., Clarke, L.,
Dahe, Q., Dasgupta, P., Dubash, N.K., Edenhofer, O., Elgizouli, I., Field, C.B., Forster, P., Friedlingstein,
P., Fuglestvedt, J., Gomez-Echeverri, L., Hallegatte, S., Hegerl, G., Howden, M., Jiang, K., Jimenez
Cisneros, B., Kattsov, V., Lee, H., Mach, K.J., Marotzke, J., Mastrandrea, M.D., Meyer, L., Minx, J.,
Mulugetta, Y., O’Brien, K., Oppenheimer, M., Pereira, J.J., Pichs-Madruga, R., Plattner, G-K., Pörtner,
H-O., Power, S.B., Preston, B., Ravindranath, N.H., Reisinger,A., Riahi, K., Rusticucci, M., Scholes,
R., Seyboth, K., Sokona, Y., Stavins, R., Stocker, T.F., Tschakert, P., van Vuuren, D. and van Ypersele,
J-P. (2014). Climate change 2014: synthesis report. Contribution of Working Groups I, II and III to the
fifth assessment report of the Intergovernmental Panel on Climate Change, Cambridge University
Press, Cambridge, UK.

Pandey, S., Gautam, R., Houweling, S., van der Gon, H.D., Sadavarte, P., Borsdorff, T., Hasekamp,
O., Landgraf, J., Tol, P., van Kempen, T., Hoogeveen, R., van Hees, R., Hamburg, S.P., Maasakkers,
J.D., Aben, I. (2019) Satellite observations reveal extreme methane leakage from a natural gas well
blowout. Proc. National Acad. Sci., 116, 26376–26381.

Parfitt, J., Barthel, M. and Macnaughton, S. (2010). Food waste within food supply chains:
quantification and potential for change to 2050. Philosophical Transactions of the Royal Society B:
Biological Sciences, 365, 3065–3081.

Petersen, S.O., Højberg, O., Poulsen, M., Schwab, C. and Eriksen, J. (2014). Methanogenic community
changes, and emissions of methane and other gases, during storage of acidified and untreated pig
slurry. Journal of Applied Microbiology, 117, 160–172.

Peterson, C.B., El Mashad, H.M., Zhao, Y., Pan, Y. and Mitloehner, F.M. (2020). Effects of SOP Lagoon
Additive on Gaseous Emissions from Stored Liquid Dairy Manure. Sustainability, 12, 1393.

Pfeiffer, A., Hepburn, C., Vogt-Schilb, A. and Caldecott, B. (2018). Committed emissions from existing
and planned power plants and asset stranding required to meet the Paris Agreement. Environmental
Research Letters, 13(5), p.054019.

Poore, J., and Nemecek, T. (2018). Reducing food’s environmental impacts through producers and
consumers. Science, 360, 987–992.

Popp, A., Lotze-Campen, H. and Bodirsky, B. (2010). Food consumption, diet shifts and associated
non-CO2 greenhouse gases from agricultural production. Global Environmental Change, 20(3):
451–462.

Popp, A., Calvin, K., Fujimori, S., Havlik, P., Humpenöder, F., Stehfest, E., Bodirsky, B.L., Dietrich,
J.P., Doelmann, J.C., Gusti, M. and Hasegawa, T. (2017). Land-use futures in the shared socio-
economic pathways. Global Environmental Change, 42, 331–345.

Porter, S.D., Reay, D.S., Bomberg, E. and Higgins, P. (2018). Avoidable food losses and associated
production-phase greenhouse gas emissions arising from application of cosmetic standards to fresh
fruit and vegetables in Europe and the UK. Journal of Cleaner Production, 201, 869–878.

Post, M.J. (2012). Cultured meat from stem cells: Challenges and prospects. Meat Science, 92(3),
297–301.

Pradhan, P., Reusser, D.E. and Kropp, J.P. (2013). Embodied greenhouse gas emissions in diets. PloS
Global Methane Assessment / REFERENCES 141

ONE, 8(5), e62228.

Prather, M.J., Holmes, C.D. and Hsu, J. (2012). Reactive greenhouse gas scenarios: Systematic
exploration of uncertainties and the role of atmospheric chemistry. Geophys. Res. Lett., 39, L09803,
doi:10.1029/2012gl051440.

Prinn, R.G., Weiss, R.F., Fraser, P.J., Simmonds, P.G., Cunnold, D.M., Alyea, F.N., O’Doherty, S.,
Salameh, P., Miller, B.R., Huang, J. and Wang, R.H.J. (2000). A history of chemically and radiatively
important gases in air deduced from ALE/GAGE/AGAGE. Journal of Geophysical Research:
Atmospheres, 105(D14), 17751–17792.

Ramanathan, V. and Xu, Y. (2010). The Copenhagen Accord for limiting global warming: Criteria,
constraints, and available avenues. Proc Natl Acad Sci USA., 107, 8055–8062.

Reeburgh, W. S. (2007). Oceanic methane biogeochemistry. Chemical Reviews, 107, 486–513.

Riahi, K., van Vuuren, D.P., Kriegler, E., Edmonds, J., O’Neill, B., Fujimori, S., Bauer, N., Calvin, K.,
Dellink, R., Fricko, O., Lutz, W., Popp, A., Crespo Cuaresma, J., Samir, K. C., Leimback, M., Jiang,
L., Kram, T., Rao, S., Emmerling, J., Ebi, K., Hasegawa, T., Havlik, P., Humpenöder, F., Da Silva, L.A.,
Smith, S., Stehfest, E., Bosetti, V., Eom, J., Gernaat, D., Masui, T., Rogelj, J., Strefler, J., Drouet, L.,
Krey, V., Luderer, G., Harmsen, M., Takahashi, K., Baumstark, L., Doelman, J., Kainuma, M., Klimont,
Z., Marangoni, G., Lotze-Campen, H., Obersteiner, M., Tabeau, A. and Tavoni, M. (2016). Shared
socioeconomic pathways: an overview. Global Environ. Change, 42, 153–168.

Robertson, A.M., Edie, R., Snare, D., Soltis, J., Field, R.A., Burkhart, M.D., Bell, C.S., Zimmerle, D. and
Murphy, S.M. (2017). Variation in Methane Emission Rates from Well Pads in Four Oil and Gas Basins
with Contrasting Production Volumes and Compositions. Environmental Science & Technology, 51,
8832–8840.

Rogelj, J., Shindell, D., Jiang, K., Fifita, S., Forster, P., Ginzburg, V., Handa, C., Kheshgi, H., Kobayashi,
S., Kriegler, E., Mundaca, L., Séférian, R. and Vilariño, M.V. (2018). Mitigation pathways compatible
with 1.5°C in the context of sustainable development. In Masson-Delmotte, V., Zhai, P., Pörtner,
H.O., Roberts, D., Skea, J., Shukla, P.R., Pirani, A., Moufouma-Okia, W., Péan, C., Pidcock, R.,
Connors, S., Matthews, J.B.R., Chen, Y., Zhou, X., Gomis, M. I., Lonnoy, E., Maycock, T., Tignor, M.
and Waterfield, T. (eds.). Global warming of 1.5°C. An IPCC Special Report on the impacts of global
warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways,
in the context of strengthening the global response to the threat of climate change, sustainable
development, and efforts to eradicate poverty. Cambridge University Press, Cambridge, UK.

Roque, B.M., Venegas, M., Kinley, R.D., De Nys, R., Duarte, T.L., Yang, X., and Kebreab, E. (2021)
Red seaweed (Asparagopsis taxiformis) supplementation reduces enteric methane by over 80 in beef
steers, PLoS One, 16, e0247820.

Samir, K.C. and Lutz, W. (2017). The human core of the shared socioeconomic pathways: Population
scenarios by age, sex and level of education for all countries to 2100. Global Env. Change, 42, 181–192.

Sarofim, M.C., Saha, S., Hawkins, M.D., Mills, D.M., Hess, J., Horton, R., Kinney, P., Schwartz,  J.
and St. Juliana, A. (2016). Temperature-Related Death and Illness, In Crimmins, A., Balbus, J.,
Gamble, J.L., Beard, C.B., Bell, J.E., Dodgen, D., Eisen, R.J., Fann, N., Hawkins, M.D., Herring, S.C.,
Jantarasami, L., Mills, D.M., Saha, S., Sarofim, M.C., Trtanj, J.and Ziska, L. (eds.). The Impacts of
Climate Change on Human Health in the United States: A Scientific Assessment. US Global Change
Research Program, Washington, DC, US.

Saunois, M., Bousquet, P., Poulter, B., Peregon, A., Ciais, P., Canadell, J.G., Dlugokencky, E.J.,
Etiope, G., Bastviken, D., Houweling, S., Janssens-Maenhout, G., Tubiello, F.N., Castaldi, S.,
Jackson, R.B., Alexe, M., Arora, V.K., Beerling, D.J., Bergamaschi, P., Blake, D.R., Brailsford, G.,
Brovkin, V., Bruhwiler, L., Crevoisier, C., Crill, P., Covey, K., Curry, C., Frankenberg, C., Gedney, N.,
Höglund-Isaksson, L., Ishizawa, M., Ito, A., Joos, F., Kim, H-S., Kleinen, T., Krummel, P., Lamarque,
J-F., Langenfelds, R., Locatelli, R., Machida, T., Maksyutov, S., McDonald, K.C., Marshall, J., Melton,
J.R., Morino, I., Naik, V., O’Doherty, S., Parmentier, F-J,W, Patra, P.K., Peng, C., Peng, S., Peters,
G.P., Pison, I., Prigent, C., Prinn, R., Ramonet, M., Riley, W.J., Saito, M., Santini, M., Schroeder, R.,
Global Methane Assessment / REFERENCES 142

Simpson, I.J., Spahni, R., Steele, P., Takizawa, A., Thornton,  B. F., Tian, H., Tohjima, Y., Viovy, N.,
Voulgarakis, A., van Weele, M., van der Werf, G.R., Weiss, R., Wiedinmyer, C., Wilton, D.J., Wiltshire,
A., Worthy, D., Wunch, D., Xu, X., Yoshida, Y., Zhang, B., Zhang, Z. and Zhu, Q. (2016). The global
methane budget 2000–2012. Earth System Science Data, 8, 2.

Saunois, M., Stavert, A.R., Poulter, B., Bousquet, P., Canadell, J.G., Jackson, R.B., Raymond, P.A.,
Dlugokencky, E.J., Houweling, S., Patra, P.K., Ciais, P., Arora, V.K., Bastviken, D., Bergamaschi,
P., Blake, D.R., Brailsford, G., Bruhwiler, L., Carlson, K.M., Carrol, M., Castaldi, S., Chandra, N.,
Crevoisier, C., Crill, P.M., Covey, K., Curry, C.L., Etiope, G., Frankenberg, C., Gedney, N., Hegglin,
M.I., Höglund-Isaksson, L., Hugelius, G., Ishizawa, M., Ito, A., Janssens-Maenhout, G., Jensen, K.M.,
Joos, F., Kleinen, T., Krummel, P.B., Langenfelds, R.L., Laruelle, G.G., Liu, L., Machida, T., Maksyutov,
S., McDonald, K. C., McNorton, J., Miller, P. A., Melton, J. R., Morino, I., Müller, J., Murguia-Flores,
F., Naik, V., Niwa, Y., Noce, S., O’Doherty, S., Parker, R.J., Peng, C., Peng, S., Peters, G.P., Prigent,
C., Prinn, R., Ramonet, M., Regnier, P., Riley, W.J., Rosentreter, J.A., Segers, A., Simpson, I.J., Shi,
H., Smith, S.J., Steele, L.P., Thornton, B.F., Tian, H., Tohjima, Y., Tubiello, F.N., Tsuruta, A., Viovy,
N., Voulgarakis, A., Weber, T.S., van Weele, M., van der Werf, G.R., Weiss, R.F., Worthy, D., Wunch,
D., Yin, Y., Yoshida, Y., Zhang, W., Zhang, Z., Zhao, Y., Zheng, B., Zhu, Q., Zhu, Q., and Zhuang, Q.
(2020). The Global Methane Budget 2000–2017, Earth Syst. Sci. Data, 12, 1561–1623.

Scarpelli, T.R., Jacob, D.J., Maasakkers, J.D., Sulprizio, M.P., Sheng, J-X., Rose, K., Romeo, L.,
Worden, J.R. and Janssens-Maenhout, G. (2020). A global gridded (0.1º×0.1º) inventory of methane
emissions from oil, gas, and coal exploitation based on national reports to the United Nations
Framework Convention on Climate Change. Earth Syst. Sci. Data, 12, 563–575.

Schneising, O., Burrows, J.P., Dickerson, R.R., Buchwitz, M., Reuter, M. and Bovensmann, H. (2014).
Remote sensing of fugitive methane emissions from oil and gas production in North American tight
geologic formations. Earth’s Future, 2, 548–558.

Schneising, O., Buchwitz, M., Reuter, M., Vanselow, S., Bovensmann, H. and Burrows, J.P. (2020).
Remote sensing of methane leakage from natural gas and petroleum systems revisited. Atmos.
Chem. Phys., 20, 9169-9281.

Schuur, E.A., McGuire, A.D., Schädel, C., Grosse, G., Harden, J.W., Hayes, D.J., Hugelius, G., Koven,
C.D., Kuhry, P., Lawrence, D.M. and Natali, S.M. (2015). Climate change and the permafrost carbon
feedback. Nature, 520, 171–179.

Schwartz, J.D., Lee, M., Kinney, P.L., Yang, S.J., Mills, D., Sarofim, M.C., Jones, R., Streeter, R., St
Juliana, A., Peers, J. and Horton, R.M. (2015). Projections of temperature-attributable premature
deaths in 209 US cities using a cluster-based Poisson approach. Environmental Health, 14.

Stockholm Environment Institute, International Institute for Sustainable Development, Overseas

Development Institute, E3G, and United Nations Environment Programme (2020). The Production
Gap Report: 2020 Special Report. http://productiongap.org/2020report.

Sekiya, T., Miyazaki, K., Ogochi, K., Sudo, K. and Takigawa, M. (2018). Global high-resolution simulations
of tropospheric nitrogen dioxide using CHASER V4. 0. Geoscientific Model Development, 11.

Sellar, A.A., Jones, C.G., Mulcahy, J., Tang, Y., Yool, A., Wiltshire, A., O’Connor, F.M., Stringer,
M., Hill, R., Palmieri, J., Woodward, S., Mora, L., Kuhlbrodt, T., Rumbold, S., Kelley, D.I., Ellis, R.,
Johnson, C.E., Walton, J., Abraham, N.L., Andrews, M.B., Andrews, T., Archibald, A.T., Berthou, S.,
Burke, E., Blockley, E., Carslaw, K., Dalvi, M., Edwards, J., Folberth, G.A., Gedney, N., Griffiths, P.T.,
Harper, A.B., Hendry, M.A., Hewitt, A.J., Johnson, B., Jones, A., Jones, C.D., Keeble, J., Liddicoat,
S., Morgenstern, O., Parker, R.J., Predoi, V., Robertson, E., Siahaan, A., Smith, R.S., Swaminathan,
R., Woodhouse, M.T., Zeng, G. and Zerroukat, M. (2019). UKESM1: Description and evaluation of the
UK Earth System Model. J. Adv. Model. Earth Syst., doi:10.1029/2019ms001739.

Seltzer, K., Shindell, D. and Malley, C. (2018). Measurement-based assessment of health burdens
from long-term ozone exposure in the United States, Europe, and China. Env. Res. Lett., 13, 104018.

Shearer, C., Tong, D., Fofrich, R. and Davis, S.J. (2020). Committed emissions of the U.S. power
sector, 2000–2018. AGU Advances, 1, e2020AV000162.
Global Methane Assessment / REFERENCES 143

Sherwood, S., Webb, M.J., Annan, J.D., Armour, K.C., Forster, P.M., Hargreaves, J.C., Hegerl, G.,
Klein, S.A., Marvel, K.D., Rohling, E.J. and Watanabe, M. (2020). An assessment of Earth’s climate
sensitivity using multiple lines of evidence. Reviews of Geophysics, e2019RG000678.

Shindell, D.T. (2012). Evaluation of the Absolute Regional Temperature Potential. Atmos. Chem.
Phys., 12, 7955–7960.

Shindell, D.T., Borgford-Parnell, N., Brauer, M., Haines, A., Kuylenstierna, J.C.I., Leonard, S.A.,
Ramanathan, V., Ravishankara, A., Amann, M. and Srivastava, L. (2017b). A climate policy pathway
for near- and long-term benefits. Science, 356, 493–494.

Shindell, D., Faluvegi, G., Kasibhatla, P. and Van Dingenen, R. (2019). Spatial patterns of crop yield
change by emitted pollutant. Earth’s Future, 7, 101–112.

Shindell, D.,  Faluvegi, G., Seltzer, K. and Shindell, C. (2018). Quantified, localized health benefits of
accelerated carbon dioxide emissions reductions. Nat. Clim. Change, 8, 291–295.

Shindell, D., Fuglestvedt, J.S. and Collins, W.J. (2017). The Social Cost of Methane: Theory and
Applications. Faraday Disc., 200, 429–451.

Shindell, D., Kuylenstierna, J.C.I., Vignati, E., Dingenen, R., Amann, M., Klimont, Z., Anenberg, S.C.,
Muller, N.., Janssens-Maenhout, G., Raes, F., Schwartz, J., Faluvegi, G., Pozzoli, L., Kupiainen, K.,
Höglund-Isaksson, L., Emberson, L., Streets, D., Ramanathan, V., Hicks, K., Oanh, K., Milly, G.,
Williams, M., Demkine, V. and Fowler, D. (2012). Simultaneously mitigating near-term climate change
and improving human health and food security. Science, 335, 183–189.

Shindell, D. and Smith, C.J. (2019). Climate and air-quality benefits of a realistic phase-out of fossil
fuels. Nature, 573, 408–411.

Shindell, D., Zhang, Y., Scott, M., Ru, M., Stark, K. and Ebi, K.L. (2020). The Effects of Heat Exposure
on Human Mortality Throughout the US. GeoHealth, 3. https://doi.org/10.1029/ 2019GH000234
(accessed 22 March 2021).

Sjögersten, S., Siegenthaler, A., Lopez, O.R., Aplin, P., Turner, B. and Gauci, V. (2020). Methane
emissions from tree stems in neotropical peatlands. The New Phytologist, 225,2, 769–781.

Skarke, A., Ruppel, C., Kodis, M., Brothers, D. and Lobecker, E.. (2014). Widespread methane leakage
from the sea floor on the northern US Atlantic margin. Nature Geosci., 7, 657–661.

Sokolov, V., VanderZaag, A., Habtewold, J., Dunfield, K., Wagner‐Riddle, C., Venkiteswaran,  J.J.
and Gordon, R. (2019). Greenhouse gas mitigation through dairy manure acidification. Journal of
environmental quality, 48, 1435–1443.

Sommer, S.G., Clough, T.J., Balaine, N., Hafner, S.D. and Cameron, K.C. (2017). Transformation
of organic matter and the emissions of methane and ammonia during storage of liquid manure as
affected by acidification. Journal of Environmental Quality, 46, 514–521.

Spadaro, J.V., Kendrovski, V. and Sanchez Martinez, G. (2018). Achieving health benefits from carbon
reductions: Manual for CaRBonH calculation tool. World Health Organization, Geneva, Switzerland.
https://core.ac.uk/reader/189891440 (accessed 9 March 2021).

Springmann, M., Clark, M., Mason-D’Croz, D., Wiebe, K., Bodirsky, B.L., Lassaletta, L., de Vries, W.,
Vermeulen, S.J., Herrero, M., Carlson, K.M., Jonell, M., Troell, M., DeClerck, F., Gordon, L.J., Zurayk,
R., Scarborough, P., Rayner, M., Loken, B., Fanzo, J., Godfray, H.C.J., Tilman, D., Rockström, J.
and Willett, W. (2018). Options for keeping the food system within environmental limits. Nature, 562,
519–525.

Springmann, M., Godfray, H.C.J., Rayner, M. and Scarborough, P. (2016). Analysis and valuation of
the health and climate change cobenefits of dietary change. Proceedings of the National Academy
of Sciences, 113(15), 4146–4151.

Stauffer, B., Fischer, G., Neftel, A. and Oeschger, H. (1985). Increase of Atmospheric Methane
Recorded in Antarctic Ice Core. Science, 1386–1388.
Global Methane Assessment / REFERENCES 144

Stehfest, E., Bouwman, L., Van Vuuren, D.P., Den Elzen, M.G.J., Eickhout, B. and Kabat, P. (2009).
Climate benefits of changing diet. Clim. Change, 95, 83–102.

Stehfest, E., van den Berg, M., Woltjer, G. and Msangi, S. (2013). Options to reduce the environmental
effects of livestock production – Comparison of two economic models. Agr. Syst., 114, 38–53.

Sterman J. D., Siegel, L. and Pooney-Varga, J.N. (2018). Does replacing coal with wood lower CO2
emissions? Dynamic lifecycle analysis of wood bioenergy. Env. Res. Lett., 13, 1–10.

Stohl, A., Aamaas, B., Amann, M., Baker, L., Bellouin, N., Berntsen, T.K., Boucher, O., Cherian, R.,
Collins, W., Daskalakis, N. and Dusinska, M., Evaluating the climate and air quality impacts of short-
lived pollutants, Atmos. Chem. Phy., 15,10529-10566, 2015.

Stolaroff J.K., Bhattacharyya S., Smith C.A., Bourcier W.L., Cameron-Smith P.J. and Aines R.D.
(2012). Review of Methane Mitigation Technologies with Application to Rapid Release of Methane
from the Arctic. Env. Sci. Tech., 46, 6455–6469.

Sudo, K., Takahashi, M., Kurokawa, J.-i., and Akimoto, H. (2002). CHASER: A global
chemical model of the troposphere 1. Model description. J. Geophys. Res., 107, ACH–7.

Sweeney, C., Karion, A., Wolter, S., Newberger, T., Guenther, D., Higgs, J.A., Andrews, A.E., Lang,
P.M., Neff, D., Dlugokencky, E., Miller, J.B., Montzka, S.A., Miller, J.B., Masarie, K.A., Biraud, S.C.,
Novelli, P.C., Crotwell, M., Crotwell, A.M., Thoning, K. and Tans, P.P. (2015).Seasonal climatology of
CO2 across North America from aircraft measurements in the NOAA/ESRL Global Greenhouse Gas
Reference Network. J. Geophys. Res. Atmos., 120, 5155–5190.

Tebaldi, C. and Lobell, D. (2018). Estimated impacts of emission reductions on wheat and maize
crops. Climatic Change, 146(3-4), 533–545.

Tilman, D. and Clark, M. (2014). Global diets link environmental sustainability and human health,
Nature, 515, 518–522.

Torero, M. (2020). Without food, there can be no exit from the pandemic. Nature, 580, 588–589.

Tuomisto, H.L. and M.J. Teixeira de Mattos. (2011). Environmental Impacts of Cultured Meat
Production. Environmental Science & Technology, 45(14), 6117–6123.

Turner, M.C., Jerrett, M., Pope, C.A., Krewski, D., Gapstur, S.M., Diver, W.R., Beckerman, B.S.,
Marshall, J.D., Su, J., Crouse, D.L. and Burnett, R.T. (2016). Long-term ozone exposure and mortality
in a large prospective study. Am. J. Respir. Crit. Care Med., 193, 1134–42.

Turner, A.J., Jacob, D.J., Benmergui, J., Wofsy, S.C., Maasakkers, J.D., Butz, A., Hasekamp, O.,
Biraud, S.C. and Duglokencky, E. (2016). A large increase in US methane emissions over the past
decade inferred from satellite data. Geophys. Res. Lett., 43, 2218–2224.

United Nations Environment Programme and World Meteorological Organization (2011). Integrated
Assessment of Black Carbon and Tropospheric Ozone. United Nations Environment Programme,
Nairobi, Kenya and World Meteorological Organization, Geneva, Switzerland.

United Nations Environment Programme (2011). Near-term Climate Protection and Clean Air Benefits:
Actions for Controlling Short-Lived Climate Forcers. United Nations Environment Programme,
Nairobi, Kenya.

United Nations Environment Programme (2017). The Emissions Gap Report 2017. United Nations
Environment Programme, Nairobi, Kenya..

United States Bureau of Labor Statistics. (2020a). Quarterly Census of Employment and Wages. U.S.
Bureau of Labor Statistics, Washington, DC, US. https://www.bls.gov/cew/datatoc.htm (accessed 9
March 2021).

United States Bureau of Labor Statistics. (2020b). Quarterly Census of Employment and Wages:
QCEW Industry Codes and Titles. U.S. Bureau of Labor Statistics, Washington, DC, US. https://data.
bls.gov/cew/doc/titles/industry/industry_titles.htm (accessed 9 March 2021).
Global Methane Assessment / REFERENCES 145

United States Environmental Protection Agency. (1997). The Benefits and Costs of the Clean Air
Act 1970-1990. Prepared for Congress by Office of Air and Radiation and Office of Policy, Planning
and Evaluation. United States Environment Protection Agency, Washington, DC, US. https://www.
epa.gov/environmental-economics/benefits-and-costs-clean-air-act-1970-1990-1997 (accessed 9
March 2021).

United States Environmental Protection Agency. (2013). Global Mitigation of Non-CO2 Greenhouse

Gases: 2010-2030. United States Environment Protection Agency, Washington, DC, US. EPA 430-
R-13-011.

United States Environmental Protection Agency. (2015). Environmental Benefits Mapping and

Analysis Program–Community Edition (BenMAP‐CE). United States Environment Protection Agency,
Washington, DC, US.

United States Environmental Protection Agency (2015a) Climate Change in the United States: Benefits
of Global Action. United States Environmental Protection Agency, Office of Atmospheric Programs,
Washington, DC, US. EPA 430-R-15-001.

United States Environmental Protection Agency. (2017). Multi‐model framework for quantitative

sectoral impacts analysis: A technical report for the Fourth National Climate Assessment. United
States Environment Protection Agency, Washington, DC, US.

United States Environmental Protection Agency (2019). Global Non-CO2 Greenhouse Gas Emission
Projections & Mitigation Potential: 2015-2050. United States Environment Protection Agency,
Washington, DC, US. EPA 430-R-19-010.

United States Global Change Research Program (2018). Climate Science Special Report: Fourth
National Climate Assessment, Volume I. Wuebbles, D.J., Fahey, D.W., Hibbard, K.A., Dokken, D.J.,
Stewart, B.C. and Maycock, T.K. (eds.). U.S. Global Change Research Program, Washington, DC, US.

Valentine, D.W., Holland, E.A. and Schimel, D.S. (1994). Ecosystem and physiological controls over
methane production in northern wetlands. J. Geophys. Res., 99, 1563–1571. 

van de Ven, D.-J., González-Eguino, M. and Arto, I. (2018). The potential of behavioural change for
climate change mitigation: a case study for the European Union. Mitigation and Adaptation Strategies
for Global Change, 23, 853–886.

Van Dingenen, R., Dentener, F.J., Raes, F., Krol, M.C., Emberson, L. and Cofala, J. (2009). The
global impact of ozone on agricultural crop yields under current and future air quality legislation.
Atmospheric Environment, 43, 604–618.

Varon, D.J., Jacob, D.J., McKeever, J., Jervis, D., Durak, B.O.A., Xia, Yan, and Huang, Y.,(2018)
Quantifying methane point sources from fine-scale satellite observations of atmospheric methane
plumes, Atmos. Meas. Tech., 11, 5673–5686.

Varon, D.J., McKeever, J., Jervis, D., Maasakkers, J.D., Pandey, S., Houweling, S., Aben, I., Scarpelli,
T. and Jacob, D.J. (2019). Satellite discovery of anomalously large methane point sources from oil/
gas production. Geophysical Research Letters, 46, 13507–13516.

Vaughn, T.L., Bell, C.C., Pickering, C.K., Schwietzke, S., Heath, G.A., Pétron, G., Zimmerle, D.J.,
Schnell, R.C. and Nummedal, D. (2018). Temporal variability largely explains top-down/bottom-up
difference in methane emission estimates from a natural gas production region. Proceedings of the
National Academy of Sciences, 115, 11712–11717.

Victor, D.G., Zaelke, D. and Ramanathan, V. (2015). Soot and short-lived pollutants provide political
opportunity. Nature Climate Change, 5, 796–798.

Vijn, S., Compart D.P., Dutta, N., Foukis, A., Hess, M., Hristov, A.N., Kalscheur, K.F., Kebreab, E.,
Nuzhdin, S.V., Price, N.N., Sun, Y., Tricarico, J.M., Turzillo, A., Weisbjerg, W.R., Yarish, C., and Kurt,
T.D. (2020) Key considerations for the Use of Seaweed to Reduce Enteric Methane Emissions From
Cattle, Front. Vet. Sci., 7, 597430.
Global Methane Assessment / REFERENCES 146

Viscusi, W.K. and Aldy, J.E. (2003). The Value of a Statistical Life: A Critical Review of Market Estimates
Throughout the World. J. Risk Uncert., 27, 5–76.

Wania, R., Melton, J.R., Hodson, E.L., Poulter, B., Ringeval, B., Spahni, R., Bohn, T., Avis, C. A.,
Chen, G., Eliseev, A.V., Hopcroft, P.O., Riley, W.J., Subin, Z.M., Tian, H., van Bodegom, P.M., Kleinen,
T., Yu, Z.C., Singarayer, J.S., Zürcher, S., Lettenmaier, D.P., Beerling, D.J., Denisov, S.N., Prigent,
C., Papa, F. and Kaplan, J.O. (2013). Present state of global wetland extent and wetland methane
modelling: methodology of a model inter-comparison project (WETCHIMP). Geosci. Model Dev., 6,
617–641.

Wang, Y., and Jacob, D.J. (1998). Anthropogenic forcing on tropospheric ozone and OH since
preindustrial times. J. Geophys. Res., 103, 31123–31135.

Wang, X. and Mauzerall, D.L. (2004). Characterizing distributions of surface ozone and its impact on
grain production in China, Japan and South Korea: 1990 and 2020. Atmospheric Environment, 38,
4383–4402.

Wang, T., Srebotnjak, T., Brownell, J. and Hsia, R.Y. (2014). Emergency department charges for
asthma-related outpatient visits by insurance status, J Health Care Poor Underserved, 25, 396–405.

Wang, W-Q. and Shao, H. (2013). Azimuth-variant signal processing in high-altitude platform passive
SAR with spaceborne/airborne transmitter. Remote Sensing, 5, 1292–1310.

Watanabe, S., Hajima, T., Sudo, K., Nagashima, T., Takemura, T., Okajima, H., Nozawa, T., Kawase,
H., Abe, M., Yokohata, T., Ise (, T., Sato, H., Kato, E., Takata, K., Emori, S. and Kawamiya, M. (2011).
MIROC-ESM 2010: Model description and basic results of CMIP5-20c3m experiments. Geoscientific
Model Development, 4, 845.

Watts, G. (2009). The Health Benefits of Tackling Climate Change: An Executive Summary. The
Lancet, London, UK.

Watts, N, Adger, W.N., Agnolucci, P., Blackstock, J., Byass, P., Cai, W., Chaytor, S., Colbourn, T.,
Collins, M., Cooper, A., Cox, P.M., Depledge, J., Drummond, P., Ekins, P., Galaz, V., Grace, D., Graham,
H., Grubb, M., Haines, A., Hamilton, I., Hunter, A., Jiang, X., Li, M., Kelman, I., Liang, L., Lott, M.,
Lowe, R., Luo, Y., Mace, G., Maslin, M., Nilsson, M., Oreszczyn, T., Pye, S., Quinn, T., Svensdotter,
M., Venevsky, S., Warner, K., Xu, B., Yang, J., Yin, Y., Yu, C., Zhang, Q., Gong, P., Montgomery, H.
and Costello, A. (2015). Health and climate change: policy responses to protect public health. The
Lancet, 386, 1861–914.

Watts, N., Amann, M., Ayeb-Karlsson, S., Belesova, K., Bouley, T., Boykoff, M., Byass, P., Cai, W.,
Campbell-Lendrum, D., Chambers, J., Cox, P.M., Daly, M., Dasandi, N., Davies, M., Depledge, M.,
Depoux, A., Dominguez-Salas, P., Drummond P., Ekins, P., Flahault, A., Frumkin, H., Georgeson, L.,
Ghanei, M., Grace, D., Graham, H., Grojsman, R., Haines, A., Hamilton, I., Hartinger, S., Johnson, A.,
Kelman, I., Kiesewetter, G., Kniveton, .C., 26, Liang, L., Lott, M., Lowe, R., Mace, G., Sewe, M.O.,
Maslin, M., Mikhaylov, S., Milner, J., Latifi, A.M., Moradi-Lakeh, M., Morrissey, K., Murray, K., Neville,
T., Nilsson, M., Oreszczyn, T., Owfi, F., Pencheon, D., Pye, S., Rabbaniha, M., Robinson, E., Rocklöv,
J., Schütte, S., Shumake-Guillemot, J., Steinbach, R., Tabatabaei, M., Wheeler, N., Wilkinson, P.,
Gong, P., Montgomery, H. and Costello, A. (2017). The Lancet Countdown on health and climate
change: from 25 years of inaction to a global transformation for public health. The Lancet, pii: S0140-
6736 (17), 32464–32469.

Weber, T., Wiseman, N.A. and Kock, A. (2019). Global ocean methane emissions dominated by
shallow coastal waters. Nat Commun., 10, 4584.

Wecht, K.J., Jacob, D.J., Frankenberg, C., Jiang, Z. and Blake, D.R. (2014). Mapping of North
America methane emissions with high spatial resolution by inversion of SCIAMACHY satellite data.
J. Geophys. Res., 119, 7741–7756.

Weinberger, K.R., Haykin, L., Eliot, M.N., Schwartz, J.D., Gasparrini, A. and Wellenius, G.A. (2017).
Projected temperature-related deaths in ten large US metropolitan areas under different climate
change scenarios. Environment International, 107, 196–204.
Global Methane Assessment / REFERENCES 147

Weiss, F. and Leip, A. (2012). Greenhouse gas emissions from the EU livestock sector: a life cycle
assessment carried out with the CAPRI model. Agriculture, Ecosystems & Environment, 149, 124–134.

Wellenius, G.A., Eliot, M.N., Bush, K.F., Holt, D., Lincoln, R.A., Smith, A.E. and Gold, J. (2017). Heat-
related morbidity and mortality in New England: Evidence for local policy. Environmental Research,
156, 845–853.

West, J.J. and Fiore, A.M. (2005). Management of tropospheric ozone by reducing methane emissions.
Environ. Sci. Technol., 39, 4685.

West, J.J., Fiore, A.M., Horowitz, L.W. and Mauzerall, D.L. (2006). Global health benefits of mitigating
ozone pollution with methane emission controls. Proc. National Acad. Sci., 103, 3988–3993.

Whalen, S.C. (2005). Biogeochemistry of Methane Exchange between Natural Wetlands and the
Atmosphere. Environ. Eng. Sci., 22, 73–94.

Willett, W., Rockström, J., Loken, B., Springmann, M., Lang, T., Vermeulen, S., Garnett, T Tilman, D.,
DeClerck, F., Wood, A., Jonell, M., Clark, M., Gordon, L.J., Fanzo, J., Hawkes, C., Zurayk, R., Rivera,
J.A., De Vries, W., Sibanda, L.M., Afshin, A., Chaudhary, A., Herrero, M., Agustina, R., Branca, F.,
Lartey, A., Fan, S., Crona, B., Fox, E., Bignet, V., Troell, M., Lindahl, T., Singh, S., Cornell, S.E., Reddy,
K.S., Narain, S., Nishtar, S. and Murray, C.J.L.

(2019). Food in the Anthropocene: the EAT–Lancet Commission on healthy diets from sustainable
food systems. The Lancet, 393, no. 10170, 447–492.

Wollenberg, E., Richards, M., Smith, P., Havlík, P., Obersteiner, M., Tubiello, F.N., Herold, M., Gerber,
P., Carter, S., Reisinger, A., van Vuuren, D.P., Dickie, A., Neufeldt, H., Sander, B.O., Wassmann, R.,
Sommer, R., Amonette, J.E., Falcucci, A., Herrero, M., Opio, C., Roman‐Cuesta, R.M., Stehfest,
E., Westhoek, H., Ortiz‐Monasterio, I., Sapkota, T., Rufino, M.C., Thornton, P.K., Verchot, L., West,
P.C., Soussana, J.‐F., Baedeker, T., Sadler, M., Vermeulen, S. and Campbell, B.M. (2016). Reducing
emissions from agriculture to meet the 2 °C target. Glob. Change Biol., 22, 3859–3864.

Worden, J., Kulawik, S., Frankenberg, C., Payne, V., Bowman, K., Cady-Peirara, K., Wecht, K., Lee,
J.-E. and Noone, D. (2012). Profiles of CH4, HDO, H2O, and N2O with improved lower tropospheric
vertical resolution from Aura TES radiances. Atmospheric Measurement Techniques, 5, 397–411.

World Health Organization (2013). Health risks of air pollution in Europe—HRAPIE project
recommendations for concentration–response functions for cost–benefit analysis of particulate
matter, ozone and nitrogen dioxide. World Health Organization, Copenhagen, Denmark.

World Health Organization (2014). Quantitative risk assessment of the effects of climate change on
selected causes of death, 2030s and 2050s. World Health Organization, Geneva, Switzerland,

World Bank. 2020. World Development Indicators. The World Bank, Washington, DC, US. https://
databank.worldbank.org/home.aspx (accessed 18 March 2020).

World Bank and Institute for Health Metrics and Evaluation (2016). The Cost of Air Pollution: Strengthening
the Economic Case for Action. The World Bank, Washington, DC, US and Institute for Health Medicine,
https://openknowledge.worldbank.org/handle/10986/25013 (accessed 10 March 2021).

Zavala-Araiza, D., Herndon, S.C., Roscioli, J.R., Yacovitch, T.I., Johnson, M.R., Tyner, D.R., Omara,
M. and Knighton, B. (2018). Methane emissions from oil and gas production sites in Alberta, Canada.
Elem Sci Anth., 6(1).

Zhang, S., Li, G., Tian, L., Guo, Q. and Pan, X. (2016). Short-term exposure to air pollution and
morbidity of COPD and asthma in East Asian area: A systematic review and meta-analysis. Environ
Res., 148, 15–23.

Zhang, Y., Gautam, R., Pandey, S., Omara, M., Maasakkers, J.D., Sadavarte, P., Lyon, D., Nesser, H.,
Sulprizio, M.P., Varon, D.J., Zhang, R., Houweling, S., Zavala-Araiza, D., Alvarez, R.A., Lorente, A.,
Hamburg, S.P., Aben, I. and Jacob, D. (2020). Quantifying methane emissions from the largest oil-
producing basin in the United States from Space. Science Advances, doi:10.1126/sciadv.aaz5120.
Global Methane Assessment / REFERENCES 148

Zhang, Z., Zimmermann, N.E., Stenke, A., Li, X., Hodson, E.L., Zhu, G., Huang, C. and Poulter,
B. (2017). Emerging role of wetland methane emissions in driving 21st century climate change.
Proceedings of the National Academy of Sciences, 114, 9647–9652.

Zhao, C., Liu, B., Piao, S., Wang, X., Lobell, D. B., Huang, Y., Huang, M., Yao, Y., Bassu, S., Ciais,
P., Durand, J-L., Elliott, J., Ewert, F., Janssens, I.A., Li, T., Lin, E., Liu, Q., Martre, P., Müller, C., Peng,
S., Peñuelas, J., Ruane, A.C., Wallach, D., Wang, T., Wu, D., Liu, Z., Zhu, Y., Zhu, Z. and Asseng, S.
(2017). Temperature increase reduces global yields of major crops in four independent estimates.
Proceedings of the National Academy of Sciences, 114, 9326–9331.

Zheng, X.Y., Ding, H., Jiang, L.N., Chen, S.W., Zheng, J.P., Qiu, M., Zhou, Y.X., Chen, Q. and Guan,
W.J. (2015). Association between Air Pollutants and Asthma Emergency Room Visits and Hospital
Admissions in Time Series Studies: A Systematic Review and Meta-Analysis. PloS One, 10, e0138146.

Zhu, Y., Purdy, K.J., Eyice, Ö. and Shen, L. (2020). Disproportionate increase in freshwater methane
emissions induced by experimental warming. Nat. Clim. Chang., 10, 685–690.

Zwiers, F.W and von Storch, H. (1995). Taking serial correlation into account in tests of the mean.
Journal of Climate, 8, 336–351.
7. ABBREVIATIONS
Global Methane Assessment / ABBREVIATIONS 150

A&E Hospital accident and emergency departments

A&EV Hospital accident and emergency department visits
ACS American Cancer Society
AFOLU Agriculture, forestry and other land use
AGAGE Advanced Global Atmospheric Gases Experiment
AGTPs Absolute global temperature potentials
AIM Asia-Pacific Integrated Model 
APHENA Air Pollution and Health: a European and North American Approach
BC Black carbon
BECCS Biofuel energy with carbon capture and storage
C centigrade
CaRBonH Carbon reduction benefits on health
CCAC Climate and Clean Air Coalition
CCS Carbon capture and storage
CEDS Community Emissions Data System
CESM Community Earth System Model
CGE Computable general equilibrium 
C 2H 2 Acetylene
CH4 Methane
CMIP Coupled Model Intercomparison Project
CNES Centre national d’études spatiales (French National Centre for Space Studies)
CO Carbon monoxide
CO2 Carbon dioxide
CO2e Carbon dioxide equivalent
DALY Disability adjusted life year
DLR Deutsches Zentrum für Luft- und Raumfahrt (German Aerospece Centre)
DRC Democratic Republic of the Congo
EDF Environmental Defense Fund
EDGAR Emissions Database for Global Atmospheric Research
e.g. exempli gratia (for example)
EJ Exajoules (1018 joules)
EPA Environment Protection Agency
ESA European Space Agency
EU European Union
FAO Food and Agriculture Organization of the United Nations
FINN Fire INventory from the National Center for Atmospheric Research 
FLW Food loss and waste
Global Methane Assessment / ABBREVIATIONS 151

FSU Former Soviet Union

GAINS Greenhouse Gas – Air Pollution Interactions and Synergies
GBD Global burden of disease
GCAM Global Change Assessment Model
GDP Gross domestic product
GFDL Geophysical Fluid Dynamics Laboratory
GFED Global Fire Emissions Database
GHG Greenhouse gas
GISS Goddard Institute of Space Studies
GLEAM Global Livestock Environmental Assessment Model
GLOBIOM Global Biosphere Management Model
GOSAT Greenhouse Gases Observing Satellite
GPW Gridded Population of the World
Gt Gigatonne (109 tonnes)
GWP Global warming potential
GWP100 Global warming potential over 100 years
HCUP Healthcare Cost and Utilization Project
HCFC Hydrochlorofluorocarbon
HFC Hydrofluorocarbon
hrs/yr Hours per year
IAM Integrated assessment model
ICD International Classification of Diseases
i.e. id est (that is)
IEA International Energy Agency
IGSD Institute for Governance & Sustainable Development
IHME Institute for Health Metrics and Evaluation
IIASA International Institute for Applied Systems Analysis
ILO International Labour Organization
IMAGE Integrated Model to Assess the Global Environment
IPCC Intergovernmental Panel on Climate Change
IPCC AR5 Intergovernmental Panel on Climate Change Fifth Assessment Report
JAMSTEC Japan Agency for Marine-Earth Science and Technology
JAXA Japan Aerospace Exploration Agency 
kg Kilogram
kt Kilotonne (1 000 tonnes)
LDAR Leak detection and repair
M Metre
Global Methane Assessment / ABBREVIATIONS 152

m2 Square metre
mm Millimetre
MAC Marginal abatement cost
MDA Maximum daily average
MDA8 Maximum daily 8-hour exposure averaged over the year

MESSAGE Model of Energy Supply Systems and their General Environmental

Impact
Met Office The United Kingdom’s Meteorological Office
MIROC Model for Interdisciplinary Research on Climate
MMM Multi-model mean
MST Mean summer temperature
Mt Million tonnes
Mt/yr Million tonnes per year
mW MilliWatts
N North
NAICS North American Industry Classification System
NASA National Aeronautics and Space Administration
NCAR National Center for Atmospheric Research
NH3 Ammonia
NHml Northern hemisphere mid-latitudes
NIES National Institute for Environmental Studies, Japan
NMVOC Non-methane volatile organic compound
NOAA National Oceanic and Atmospheric Administration
NOx Nitrogen oxides
N 2O Nitrous oxide
NTCF Near-term climate forcers
NSO Netherlands Space Organisation
O3 Ozone
OC Organic carbon
OECD Organisation for Economic Co-operation and Development
OH Hydroxide
PBL Netherlands Environmental Assessment Agency
PM Particulate matter
ppb Parts per billion
ppbv Parts per billion by volume
QFED Quick Fire Emissions Dataset
RCP Representative concentration pathway
Global Methane Assessment / ABBREVIATIONS 153

REMIND REgional Model of INvestments and Development 

RR Relative Risk
S South

SCIAMACHY Scanning Imaging Absorption Spectrometer for Atmospheric

Cartography
SDG Sustainable Development Goal
SEI Stockholm Environment Institute
SHext Southern hemisphere extratropics
SLCP Short-lived climate pollutant
SO2 Sulphur dioxide
SSP Shared socioeconomic pathway
t Tonne
Tg Teragram (1012 grams)
TMREL Theoretical minimum risk exposure level
TROPOMI Tropospheric Monitoring Instrument
UK United Kingdom
UKESM United Kingdom Earth System Model
UN United Nations
UNEP United Nations Environment Program
UNFCCC United Nations Framework Convention on Climate Change
US United States of America
US$ United States dollar
USDA United States Department of Agriculture
US EPA United States Environmental Protection Agency
UV Ultraviolet
VOC Volatile organic compound
VSL Value of statistical life
WA Work availability
WBGT Wet bulb globe temperature
WITCH World Induced Technical Change Hybrid model
WL Work loss
WMO World Meteorological Organization
WTP Willingness to pay
yr Year

Note Throughout this assessment 1 billion = 1 000 000 000 (109)

 8. ANNEX –
ADDITIONAL
FIGURES AND
TABLES
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 155

Figures A1 and A2 and Tables A1–A3 report values from the modelling performed in support of this
assessment described in Chapter 3.
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 156

Figure A1 Average temperature responses to methane increase from one-half present methane to the present value,
along with the ozone and stratospheric water vapour responses to that methane increase, for the indicated seasons,
degrees Centigrade

Note: The multi-model mean is shown in the lower right map for each season, with values included
only for areas in which the value is statistically significant – grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 157

Figure A2 As Figure A1 but showing precipitation for solstice seasons only, millimetres

Note: The multi-model mean is shown in the lower right map for each season, with values included
only for areas in which the value is statistically significant – grey indicates no quantification.

Source: UNEP and CCAC

Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 158

Table A1 Total and per person respiratory deaths per 10 Mt of methane for all countries with non-zero values

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE

India 2 045 1 169 2 863 Lesotho 5 4 7

China 1 419 948 1 831 Korea, Dem People’s Rep. 5 3 6
United States 428 287 553 Eritrea 4 2 6
Japan 251 176 316 United Kingdom 4 2 5
Brazil 160 100 216 India 4 2 5
Pakistan 159 98 215 Portugal 3 2 4
United Kingdom 157 102 207 São Tome and Principe 3 2 4
Indonesia 154 86 219 Guinea-Bissau 3 2 5
Germany 121 79 160 Zimbabwe 3 2 4
Russia 109 71 143 Denmark 3 2 4
Philippines 101 64 137 Belgium 3 2 4
Spain 89 59 116 Somalia 3 2 4
Bangladesh 82 51 111 Myanmar 3 2 4
Egypt 82 57 104 Nepal 3 2 4
Mexico 77 50 100 Argentina 3 2 4
Italy 74 49 96 Swaziland 3 2 4
Myanmar 72 45 97 Spain 3 2 4
France 69 45 92 Japan 3 2 3
Korea, Dem Peoples Rep. 64 45 80 Netherlands 3 2 4
Argentina 63 40 86 Gambia 3 1 4
Turkey 61 42 79 Chad 3 1 4
Viet Nam 61 38 82 Cuba 3 2 3
Ethiopia 58 35 80 Haiti 3 2 3
Thailand 52 31 72 Sri Lanka 3 1 4
South Africa 51 34 67 Philippines 2 2 3
Nigeria 42 22 61 Afghanistan 2 2 3
Korea, Rep. 39 27 49 Guinea 2 1 3
Iran, Islamic Rep. 38 27 47 Senegal 2 1 4
Poland 37 24 50 Djibouti 2 1 3
Canada 35 24 46 Grenada 2 2 3
DRC 34 21 46 Central African Rep. 2 1 3
Ukraine 33 21 44 Pakistan 2 1 3
Nepal 32 20 42 Niger 2 1 3
Netherlands 29 19 39 Mozambique 2 1 3
Sri Lanka 28 15 40 Barbados 2 1 3
Peru 27 15 37 Norway 2 1 3
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 159

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Malaysia 26 15 36 Uruguay 2 1 3
Kenya 25 15 36 United States 2 1 3
Tanzania 25 15 35 Greece 2 1 3
Portugal 23 15 31 Egypt 2 1 3
Morocco and Western
23 15 31 South Africa 2 1 3
Sahara
Afghanistan 23 16 28 Cape Verde 2 1 3
Belgium 23 14 30 Germany 2 1 3
Australia 21 13 28 Israel 2 1 3
Romania 20 13 26 St. Lucia 2 1 3
Cuba 19 12 25 Malta 2 1 3
Sudan 18 11 25 Namibia 2 1 3
Mozambique 18 11 24 Mauritania 2 1 3
Ghana 18 9 26 Ireland 2 1 3
Greece 17 11 23 Burundi 2 1 3
Puerto Rico and the Virgin
Algeria 17 11 23 2 1 3
Islands
Saudi Arabia 17 11 21 Ethiopia 2 1 3
Chile 16 11 21 Benin 2 1 3
Kazakhstan 15 10 19 Ghana 2 1 3
Zimbabwe 14 9 19 Cameroon 2 1 3
Colombia 14 8 19 Sweden 2 1 2
Uzbekistan 13 9 17 Malawi 2 1 3
Uganda 13 7 19 Peru 2 1 3
Venezuela 13 8 18 Malaysia 2 1 3
Cameroon 13 7 17 Georgia 2 1 2
Yemen 12 8 16 Armenia 2 1 2
Niger 12 6 17 Comoros 2 1 2
Czechia 12 8 15 Kenya 2 1 3
Sweden 12 8 15 El Salvador 2 1 2
Angola 12 7 15 Burkina Faso 2 1 3
Côte d’Ivoire 11 6 16 Italy 2 1 2
Denmark 11 7 15 Luxembourg 2 1 2
Somalia 11 7 15 Bahamas 2 1 2
Senegal 11 6 16 Sierra Leone 2 1 2
Hungary 10 7 14 China 2 1 2
Haiti 10 6 13 Zambia 2 1 2
Cambodia 10 6 13 Jamaica 2 1 2
Serbia and Montenegro 9 6 12 Tanzania 2 1 2
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 160

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Chad 9 5 13 France 2 1 2
Madagascar 9 0 0 Mali 2 1 3
Guinea 9 5 13 Kazakhstan 2 1 2
Israel 9 6 11 Côte d’Ivoire 2 1 2
Malawi 9 5 12 Bolivia 2 1 2
Burkina Faso 8 4 13 Timor-Leste 2 1 2
Guatemala 8 5 11 Bhutan 2 1 2
Mali 8 4 11 DRC 2 1 2
Syrian Arab Rep. 7 5 10 Czechia 2 1 2
Bulgaria 7 5 10 Chile 2 1 2
Switzerland 7 5 10 Botswana 2 1 2
Norway 7 5 10 Cyprus 2 1 2
Bolivia 7 5 9 Hungary 2 1 2
Zambia 7 5 10 St.Vincent and
the Grenadines 2 1 2
Austria 7 5 9 Turkey 2 1 2
Tunisia 7 5 9 Yemen 2 1 2
Ecuador 6 4 9 Cambodia 2 1 2
Benin 6 3 9 Angola 2 1 2
Dominican Rep. 6 4 8 Togo 2 1 2
Eritrea 6 4 8 Serbia and Montenegro 2 1 2
Ireland 6 4 8 Congo 2 1 2
Burundi 6 3 8 Iceland 2 1 2
Rwanda 5 3 8 Canada 2 1 2
Azerbaijan 5 4 7 Romania 2 1 2
Iraq 5 4 7 Guatemala 1 1 2
Singapore 5 3 7 Brazil 1 1 2
El Salvador 5 3 7 Bulgaria 1 1 2
Belarus 5 3 6 Poland 1 1 2
Georgia 4 3 6 Rwanda 1 1 2
Slovakia 4 3 6 Croatia 1 1 2
United Arab Emirates 4 3 6 Australia 1 1 2
Puerto Rico and the Virgin Morocco and Western
4 3 6 1 1 2
Islands Sahara
Uruguay 4 3 6 Antigua and Barbuda 1 1 2
Croatia 4 3 5 Lao People’s Dem Rep. 1 1 2
Tajikistan 4 3 5 Singapore 1 1 2
Lebanon 4 2 5 Uganda 1 1 2
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 161

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Sierra Leone 4 2 5 Liberia 1 1 2
Lesotho 4 2 5 Lebanon 1 1 2
Togo 4 2 5 Tajikistan 1 1 2
Honduras 3 2 5 Switzerland 1 1 2
Finland 3 2 5 Mexico 1 1 2
Lao People’s Dem Rep. 3 2 5 Bosnia and Herzegovina 1 1 2
Libyan Arab Jamahiriya 3 2 4 Slovenia 1 1 2
Central African Rep. 3 2 4 Dominican Rep. 1 1 2
Bosnia and Herzegovina 3 2 4 Viet Nam 1 1 2
Armenia 3 2 4 Indonesia 1 1 2
Kyrgyzstan 3 2 4 Gabon 1 1 2
Mauritania 3 2 4 Belize 1 1 2
Moldova 3 2 4 Slovakia 1 1 2
Jordan 2 2 3 Thailand 1 1 2
Congo 2 1 3 Kyrgyzstan 1 1 2
Jamaica 2 2 3 Madagascar 1 1 2
Costa Rica 2 1 3 Bangladesh 1 1 2
Lithuania 2 1 3 Tunisia 1 1 2
Table A2 Total and per person cardiovascular deaths per 10 Mt of methane for all countries with non-zero values
Nicaragua 2 1 3 Austria 1 1 2

Liberia 2 1 3 Lithuania 1 1 2
Slovenia 2 1 2 Russia 1 1 2
Paraguay 2 1 2 Sudan 1 1 2
Panama 2 1 2 Korea, Rep. 1 1 1
Guinea-Bissau 2 1 2 Libyan Arab Jamahiriya 1 1 1
Namibia 2 1 2 Syria 1 1 1
Albania 2 1 2 Moldova 1 1 1
Gambia 1 1 2 Ukraine 1 1 1
Botswana 1 1 2 Azerbaijan 1 1 1
Turkmenistan 1 1 2 Saudi Arabia 1 1 1
Latvia 1 1 2 Trinidad and Tobago 1 1 2
Oman 1 1 2 Honduras 1 1 1
Cyprus 1 1 1 Uzbekistan 1 1 1
Swaziland 1 1 1 Iran, Islamic Rep. 1 1 1
North Macedonia 1 1 1 Algeria 1 1 1
Gabon 1 1 1 Finland 1 1 1
Trinidad and Tobago 1 0 1 Equatorial Guinea 1 1 1
Djibouti 1 1 1 Albania 1 1 1
Estonia 1 0 1 Latvia 1 1 1
Kuwait 1 0 1 Costa Rica 1 1 1
Mongolia 1 0 1 Panama 1 1 1
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 162

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Venezuela 1 1 1
Maldives 1 1 1
Estonia 1 1 1
Ecuador 1 0 1
Nigeria 1 0 1
Nicaragua 1 1 1
United Arab Emirates 1 1 1
Belarus 1 0 1
Jordan 1 0 1
North Macedonia 1 0 1
Brunei Darussalam 1 0 1
Paraguay 1 0 1
Oman 1 0 1
Turkmenistan 1 0 1
Colombia 1 0 1
Mongolia 1 0 1

Table A2 Total and per person cardiovascular deaths per 10 Mt of methane for all countries with non-zero values

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
China 1 566 502 2 522 Ukraine 7 2 11
India 1 122 297 1 844 Georgia 5 2 9
Russia 379 122 624 Uzbekistan 5 2 8
United States 333 108 540 Bulgaria 5 2 9
Pakistan 225 65 365 Egypt 5 2 8
Indonesia 219 57 368 Serbia and Montenegro 5 2 8
Egypt 194 64 310 Belarus 5 2 8
Ukraine 193 61 318 Lithuania 5 1 8
Japan 148 50 236 Latvia 4 1 7
Germany 133 42 218 Afghanistan 4 1 7
Brazil 95 29 159 Russia 4 1 7
Bangladesh 93 27 151 Romania 4 1 7
Italy 88 28 142 Armenia 4 1 6
United Kingdom 76 24 125 Azerbaijan 4 1 6
Iran, Islamic Rep. 76 25 120 Kyrgyzstan 4 1 6
Uzbekistan 70 23 112 Turkmenistan 4 1 6
Viet Nam 66 20 108 Bosnia and Herzegovina 4 1 6
Morocco and Western
Poland 65 21 108 4 1 6
Sahara
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 163

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Mexico 63 20 103 Moldova 4 1 6
Morocco and Western
62 19 101 Croatia 4 1 6
Sahara
Philippines 61 19 102 Kazakhstan 3 1 6
France 60 19 99 Yemen 3 1 5
Turkey 59 19 96 Pakistan 3 1 5
Spain 56 18 90 Hungary 3 1 5
Romania 53 17 87 Syria 3 1 5
Algeria 41 13 67 Albania 3 1 5
Afghanistan 40 13 63 North Macedonia 3 1 5
Sudan 39 11 65 Lebanon 3 1 5
Korea, Dem People’s Rep. 34 11 54 Tajikistan 3 1 5
Kazakhstan 30 10 48 Slovakia 3 1 5
Belarus 29 9 48 Czechia 3 1 5
Serbia and Montenegro 29 9 48 Tunisia 3 1 5
Argentina 28 9 48 Greece 3 1 5
Canada 28 9 46 Malta 3 1 4
Thailand 28 8 46 Poland 3 1 4
Saudi Arabia 28 9 44 Haiti 3 1 4
Nigeria 27 7 46 Estonia 3 1 4
Ethiopia 27 8 44 Sudan 3 1 4
Bulgaria 26 8 43 Lesotho 2 1 4
Korea, Rep. 26 9 41 Korea, Dem People’s Rep. 2 1 4
Yemen 25 8 41 Mongolia 2 1 4
South Africa 23 7 38 Cuba 2 1 4
Myanmar 22 7 36 Germany 2 1 4
Greece 22 7 35 Algeria 2 1 4
Hungary 22 7 35 Dominican Rep. 2 1 4
Iraq 21 7 33 Jamaica 2 1 4
Syria 21 7 33 Libyan Arab Jamahiriya 2 1 4
Czechia 21 7 34 Portugal 2 1 4
DRC 19 6 32 Italy 2 1 3
Azerbaijan 19 6 31 Eritrea 2 1 3
Nepal 19 6 30 Bahamas 2 1 3
Sri Lanka 18 5 30 Iran, Islamic Rep. 2 1 3
Cuba 17 5 28 Austria 2 1 3
Venezuela 17 5 28 India 2 1 3
Tunisia 17 5 27 Sweden 2 1 3
Netherlands 16 5 27 Finland 2 1 3
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 164

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Portugal 16 5 26 Grenada 2 1 3
Australia 15 5 25 China 2 1 3
St. Vincent and the
Malaysia 15 4 25 2 1 3
Grenadines
Belgium 14 4 22 Belgium 2 1 3
Georgia 13 4 21 Guinea-Bissau 2 0 3
Tanzania 12 4 21 United Kingdom 2 1 3
Sweden 12 4 20 Indonesia 2 0 3
Mozambique 12 4 19 Iraq 2 1 3
Austria 11 4 19 Trinidad and Tobago 2 0 3
Kenya 11 3 18 Saudi Arabia 2 1 3
Dominican Rep. 10 3 17 United States 2 1 3
Slovakia 10 3 17 Barbados 2 1 3
Haiti 10 3 17 Slovenia 2 1 3
Croatia 10 3 16 Spain 2 1 3
Colombia 10 3 17 St. Lucia 2 1 3
Ghana 10 2 17 Cyprus 2 1 3
Chile 10 3 16 Nepal 2 1 3
Kyrgyzstan 9 3 15 Somalia 2 1 3
Moldova 9 3 15 Israel 2 1 3
Lithuania 9 3 15 Japan 2 1 3
Tajikistan 9 3 14 Sri Lanka 2 0 3
Turkmenistan 9 3 14 Gambia 2 0 3
Bosnia and Herzegovina 9 3 15 Djibouti 2 0 3
Lebanon 8 3 13 Cape Verde 2 0 3
Switzerland 8 3 13 Turkey 2 0 2
Peru 8 2 13 Timor-Leste 1 0 3
Côte d’Ivoire 7 2 12 Denmark 1 0 2
Finland 7 2 11 Netherlands 1 0 2
Israel 7 2 11 Philippines 1 0 2
Angola 6 2 11 Mozambique 1 0 2
Armenia 6 2 10 Switzerland 1 0 2
Uganda 6 1 11 Luxembourg 1 0 2
Somalia 6 2 10 France 1 0 2
Libyan Arab Jamahiriya 6 2 10 Norway 1 0 2
Cambodia 6 2 10 Central African Rep. 1 0 2
Zimbabwe 6 2 10 Oman 1 0 2
Latvia 6 2 10 Viet Nam 1 0 2
Senegal 6 1 10 Bangladesh 1 0 2
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 165

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Denmark 5 2 9 Jordan 1 0 2
Albania 5 2 9 Lao People’s Dem Rep. 1 0 2
Niger 5 1 9 Iceland 1 0 2
Madagascar 5 1 9 Swaziland 1 0 2
Cameroon 5 1 8 Ireland 1 0 2
Puerto Rico and the Virgin
Guinea 5 1 8 1 0 2
Islands
United Arab Emirates 5 1 8 Uruguay 1 0 2
Norway 5 1 8 Antigua and Barbuda 1 0 2
Mali 5 1 8 Guinea 1 0 2
Jordan 5 1 7 Senegal 1 0 2
Burkina Faso 5 1 8 Zimbabwe 1 0 2
Chad 4 1 7 Argentina 1 0 2
Honduras 4 1 7 Chad 1 0 2
Malawi 4 1 7 Honduras 1 0 2
Guatemala 4 1 7 El Salvador 1 0 2
North Macedonia 4 1 6 Canada 1 0 2
Ireland 4 1 6 Venezuela 1 0 2
Zambia 4 1 6 Mauritania 1 0 2
Bolivia 3 1 6 Mexico 1 0 2
Ecuador 3 1 6 São Tome and Principe 1 0 2
El Salvador 3 1 6 Ghana 1 0 2
Lao People’s Dem Rep. 3 1 5 Malaysia 1 0 2
Jamaica 3 1 5 Bhutan 1 0 2
Mongolia 3 1 5 Niger 1 0 2
Eritrea 3 1 5 Australia 1 0 2
Puerto Rico and the Virgin
3 1 5 Sierra Leone 1 0 2
Islands
Benin 3 1 5 Namibia 1 0 2
Singapore 3 1 5 Mali 1 0 2
Oman 3 1 4 Comoros 1 0 2
Burundi 3 1 4 Côte d’Ivoire 1 0 2
Uruguay 3 1 4 South Africa 1 0 2
Slovenia 2 1 4 Chile 1 0 2
Estonia 2 1 4 Cambodia 1 0 2
Sierra Leone 2 1 4 Burundi 1 0 2
Togo 2 0 4 Burkina Faso 1 0 2
Nicaragua 2 1 3 DRC 1 0 2
Costa Rica 2 1 3 Ethiopia 1 0 2
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 166

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Central African Rep. 2 1 3 Liberia 1 0 2
Rwanda 2 0 3 Myanmar 1 0 1
Paraguay 2 0 3 Togo 1 0 2
Lesotho 2 1 3 Malawi 1 0 1
Mauritania 2 0 3 Benin 1 0 2
Panama 1 0 2 Brazil 1 0 1
Congo 1 0 2 Angola 1 0 1
Trinidad and Tobago 1 0 2 Congo 1 0 1
Kuwait 1 0 2 Botswana 1 0 1
Liberia 1 0 2 Zambia 1 0 1
Cyprus 1 0 2 United Arab Emirates 1 0 1
Guinea-Bissau 1 0 2 Tanzania 1 0 1
Gambia 1 0 1 Bolivia 1 0 1
Namibia 1 0 1 Nicaragua 1 0 1
Botswana 1 0 1 Singapore 1 0 1
Malta 1 0 1 Kenya 1 0 1
Korea, Rep. 1 0 1
Costa Rica 1 0 1
Belize 1 0 1
Cameroon 1 0 1
Maldives 1 0 1
Panama 1 0 1
Guatemala 1 0 1
Gabon 1 0 1
Madagascar 1 0 1
Paraguay 1 0 1
Thailand 1 0 1
Uganda 1 0 1
Bahrain 1 0 1
Kuwait 1 0 1
Peru 1 0 1
Nigeria 1 0 1
Rwanda 1 0 1
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 167

Table A3 Total and per person respiratory + cardiovascular deaths per 10 Mt of methane for all countries with non-
zero values
PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
India 3 167 472 5 567 Ukraine 8 2 13
China 2 984 729 5 003 Lesotho 8 2 13
United States 761 189 1 284 Egypt 7 2 12
Russian 488 117 839 Georgia 7 2 12
Japan 399 108 659 Korea, Dem People’s Rep. 7 2 11
Pakistan 384 74 659 Afghanistan 7 2 11
Indonesia 373 52 671 Bulgaria 7 2 12
Egypt 276 72 457 Serbia and Montenegro 6 2 11
Brazil 255 54 447 Uzbekistan 6 2 10
Germany 254 60 436 Eritrea 6 1 11
United Kingdom 233 55 400 Lithuania 6 1 10
Ukraine 226 51 390 Armenia 6 2 10
Bangladesh 176 34 300 United Kingdom 6 1 10
Philippines 162 35 285 Pakistan 6 1 10
Italy 162 39 273 Belarus 6 1 10
Spain 145 35 245 Romania 6 1 10
Mexico 140 33 237 India 5 1 10
France 130 29 223 Portugal 5 1 9
Viet Nam 127 27 219 Russia 5 1 9
Turkey 121 31 203 Latvia 5 1 9
Morocco and Western
Iran, Islamic Rep. 114 31 186 5 1 9
Sahara
Poland 103 23 178 Bosnia and Herzegovina 5 1 9
Korea, Dem People’s
97 27 160 Haiti 5 1 9
Rep.
Myanmar 94 20 163 Kyrgyzstan 5 1 8
Argentina 92 20 161 Kazakhstan 5 1 9
Morocco and Western
Sahara 85 19 145 Greece 5 1 8
Ethiopia 84 16 149 Guinea-Bissau 5 0 9
Uzbekistan 83 22 138 Azerbaijan 5 1 8
Thailand 80 14 141 Croatia 5 1 8
South Africa 75 18 127 Belgium 5 1 9
Romania 73 17 125 Moldova 5 1 8
Nigeria 69 8 128 Yemen 5 1 8
Korea, Rep. 65 17 107 Cuba 5 1 9
Canada 64 16 108 Hungary 5 1 8
Afghanistan 62 17 102 Malta 5 1 8
Algeria 58 14 99 Somalia 5 1 8
Sudan 57 10 102 Denmark 5 1 8
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 168

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
DRC 54 10 94 Nepal 5 1 8
Nepal 50 11 84 Czechia 4 1 8
Netherlands 45 10 79 Germany 4 1 8
Sri Lanka 45 6 82 Spain 4 1 8
Saudi Arabia 44 11 74 Turkmenistan 4 1 7
Kazakhstan 44 11 75 Zimbabwe 4 1 8
Malaysia 40 6 72 Lebanon 4 1 7
Portugal 39 9 67 Tajikistan 4 1 7
Greece 39 9 66 Japan 4 1 7
Serbia and Montenegro 39 9 66 Syria 4 1 7
Tanzania 38 7 67 São Tome and Principe 4 1 7
Yemen 37 9 64 Grenada 4 1 7
Kenya 36 6 66 Slovakia 4 1 7
Belgium 36 8 63 Gambia 4 0 8
Cuba 36 7 63 Netherlands 4 1 7
Australia 36 8 63 Argentina 4 1 7
Peru 34 5 62 Swaziland 4 1 7
Belarus 34 8 59 Sri Lanka 4 1 7
Bulgaria 34 8 58 Poland 4 1 7
Czechia 32 7 56 Albania 4 1 7
Hungary 32 7 55 Tunisia 4 1 7
Venezuela 30 6 53 Barbados 4 1 7
Mozambique 29 7 51 Philippines 4 1 7
Syria 28 7 47 United States 4 1 7
Ghana 28 2 52 Djibouti 4 1 7
Iraq 26 7 43 Jamaica 4 1 7
Chile 26 6 44 Myanmar 4 1 7
Azerbaijan 24 6 41 Chad 4 1 7
Tunisia 24 5 40 St. Lucia 4 1 7
Sweden 24 6 41 Sweden 4 1 7
Colombia 24 4 43 Italy 4 1 6
Haiti 20 4 35 Mozambique 4 1 7
Zimbabwe 20 5 34 Central African Rep. 4 1 7
Uganda 19 2 36 Guinea 4 1 7
Austria 18 5 31 Israel 4 1 6
Côte d’Ivoire 18 3 33 Senegal 4 0 7
Angola 18 4 31 North Macedonia 4 1 6
Cameroon 18 3 31 Bahamas 4 1 6
Georgia 17 5 29 Sudan 4 1 7
Niger 17 2 32 Cape Verde 4 1 6
Somalia 17 3 30 Norway 4 1 6
Denmark 17 4 29 Dominican Rep. 4 1 6
Dominican Rep. 16 3 29 Uruguay 4 1 6
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 169

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Senegal 16 2 30 China 4 1 6
St. Vincent and the
Switzerland 16 4 27 3 1 6
Grenadines
Cambodia 16 3 27 Estonia 3 1 6
Israel 15 3 26 Niger 3 0 6
Slovakia 15 3 26 Ireland 3 1 6
Puerto Rico and the Virgin
Madagascar 14 0 0 3 1 6
Islands
Croatia 14 3 24 Libyan Arab Jamahiriya 3 1 6
Guinea 14 2 25 Cyprus 3 1 5
Chad 13 2 24 Algeria 3 1 6
Tajikistan 13 4 22 Mauritania 3 1 6
Burkina Faso 13 1 25 Luxembourg 3 1 6
Malawi 13 3 22 Timor-Leste 3 1 6
Guatemala 12 3 21 South Africa 3 1 5
Mali 12 1 23 Austria 3 1 5
Kyrgyzstan 12 3 20 France 3 1 5
Lebanon 12 3 20 Namibia 3 1 5
Moldova 12 3 21 Turkey 3 1 5
Bosnia and Herzegovina 12 3 20 Indonesia 3 0 6
Norway 12 3 20 Slovenia 3 1 5
Lithuania 11 3 20 El Salvador 3 1 5
Zambia 11 2 19 Ghana 3 0 6
Bolivia 11 3 18 Burundi 3 0 5
Turkmenistan 10 3 17 Iran, Islamic Rep. 3 1 5
Finland 10 2 18 Malaysia 3 0 5
Ecuador 10 2 18 Ethiopia 3 1 5
Ireland 10 2 16 Iceland 3 1 5
Armenia 9 3 16 Mongolia 3 1 5
Libyan Arab Jamahiriya 9 2 16 Trinidad and Tobago 3 0 5
United Arab Emirates 9 2 15 Saudi Arabia 3 1 5
Benin 9 1 17 Benin 3 0 5
Eritrea 9 2 16 Comoros 3 1 5
El Salvador 8 2 14 Finland 3 1 5
Burundi 8 1 15 Switzerland 3 1 5
Singapore 8 1 14 Lao People’s Dem Rep. 3 1 5
Honduras 8 2 13 Malawi 3 1 5
Puerto Rico and the
7 2 12 Canada 3 1 5
Virgin Islands
Rwanda 7 1 13 Antigua and Barbuda 3 1 5
Jordan 7 2 12 Sierra Leone 3 0 5
Latvia 7 2 12 Bhutan 3 1 4
Albania 7 2 12 Mali 3 0 5
Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 170

PER
COUNTRY TOTAL LOW HIGH COUNTRY MILLION LOW HIGH
PEOPLE
Uruguay 7 1 12 Viet Nam 3 1 5
Lao People’s Dem Rep. 7 1 11 Côte d’Ivoire 3 0 5
Sierra Leone 6 1 10 Burkina Faso 3 0 5
Togo 6 0 11 Cameroon 3 0 5
Jamaica 5 1 9 Chile 3 1 4
Lesotho 5 1 9 Bangladesh 3 0 4
Central African Rep. 5 1 8 Kenya 3 0 5
North Macedonia 5 1 8 DRC 3 0 4
Slovenia 4 1 7 Cambodia 3 1 4
Mauritania 4 1 8 Zambia 3 1 4
Costa Rica 4 1 7 Tanzania 3 0 4
Nicaragua 4 1 7 Bolivia 2 1 4
Oman 4 1 7 Australia 2 1 4
Congo 4 1 7 Botswana 2 1 4
Mongolia 4 1 6 Togo 2 0 5
Paraguay 4 1 6 Angola 2 1 4
Liberia 3 1 5 Mexico 2 1 4
Panama 3 1 6 Congo 2 0 4
Estonia 3 1 5 Peru 2 0 4
Guinea-Bissau 3 0 5 Brazil 2 1 4
Namibia 2 1 4 Honduras 2 0 4
Gambia 2 0 4 Liberia 2 0 4
Trinidad and Tobago 2 0 4 Iraq 2 1 4
Cyprus 2 1 4 Guatemala 2 0 4
Kuwait 2 0 4 Singapore 2 0 4
Botswana 2 1 4 Jordan 2 1 4
Swaziland 2 0 3 Venezuela 2 0 4
Gabon 1 0 3 Uganda 2 0 4
Djibouti 1 0 2 Belize 2 0 4
Malta 1 0 2 Oman 2 0 3
Luxembourg 1 0 2 Rwanda 2 0 4
Timor-Leste 1 0 2 Gabon 2 0 4
Bhutan 1 0 1 Korea, Rep. 2 1 3
Qatar 1 0 1 Madagascar 2 0 0
Bahamas 1 0 1 Thailand 2 0 3
Bahrain 1 0 1 Costa Rica 2 0 3
Cape Verde 1 0 1 Panama 2 0 3
Barbados 1 0 1 Maldives 2 0 3
Comoros 1 0 1 United Arab Emirates 2 0 3
Iceland 1 0 1 Nicaragua 2 0 3
Mauritius 1 0 0 Ecuador 1 0 2
Equatorial Guinea 1 0 1 Equatorial Guinea 1 0 2
Paraguay 1 0 2
 Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 171

Table A4 Methane measurement techniques, the methods used by each technique and the advantages and
disadvantages of each method

Source: Methane Emission Measurement and Monitoring Methods, National Academies of Sciences,
Engineering, and Medicine (2018)

TECHNIQUE METHOD ADVANTAGES DISADVANTAGES

Measurement Measures
of emissions total methane
Ground-based from fixed points emissions from Limited number of methane sources are emitted as
measurements based on flow individual point point sources.
rate and methane sources (e.g.,
composition. stacks, animals).
Captures temporal
Labour intensive to quantify spatial and temporal
trends if deployed
variability (requires a large number of individual
for extended time
measurements to capture variability).
periods.
Often limited to measurements from normaloperations
or where there are no safety concerns.
Release of tracer Measures
gas (C2H2, N2O) total methane Difficult to isolate individual sources within source area
External tracer
at known rate emissions from depending on layout and meteorological conditions.
from source area. source area.
Measurement
of methane Measures
and tracer complex sources
concentrations or quantifies the Appropriate meteorological conditions are necessary
across well-mixed uncertainty in the for technique to work properly.
downwind plumes emission estimate
to derive emission (multiple tracers).
rate.
Vulnerable to bias if the locations of tracer release differ
significantly from the location of methane release.
Labour intensive to measure the spatial and temporal
variability of emissions over many sources.
Difficult to isolate various sources within the source
area depending on source layout and meteorological
conditions.
Remotely
Multiple vertical
measures
measurements
total methane
of atmospheric
emissions from
Facility-scale methane and Generally, cannot isolate individual sources within
a source area/
in situ aircraft wind-speed source area unless useful source-specific tracers can
facility regardless
measurements gradients above be co-quantified.
of the operational
a source area
status or safety
to derive an
conditions at the
emission rate.
facility.
Captures
Appropriate meteorological conditions (sufficient
temporal trends
vertical mixing of surface emissions at flyover
with repeated
elevations, typically midday conditions) are necessary.
overflights.
Requires multiple flights to capture temporal trends in
emissions.
Generally limited to higher-emitting sources (lower
detection limits are much higher than point-source
techniques).
Labour intensive to measure the spatial and temporal
variability of emissions over many sources.
 Global Methane Assessment / ANNEX – ADDITIONAL FIGURES AND TABLES 172

Methane
by infrared
Sparse spatial coverage, potential small-sensitivity
Towers spectrometry at High precision.
footprint.
precise infrared
wavelengths.
Time-series
measurements of Consistent
concentrations, measurements
Methods are not fully developed.
analyzed by eddy across multiple
covariance or by sites.
inverse modeling.
Challenging to apply to individual facilities and
Long time series.
distinguish confounding sources.
Ability to target
specific emission
Measurements
Aircraft mass source regions
upwind and Limited spatial coverage; temporal coverage limited to
balance and obtain
downwind of a snapshot.
measurements vertical profiles
source region.
of methane
concentrations.
Analyzed
using simple
High time- flow-through
Challenging to account for transient plumes through the
resolution models and/or
box.
instruments. sophisticated
inversion
modelling.
Infrared
spectrometry at Labour intensive to measure the spatial and temporal
precise infrared variability of emissions.
wavelengths.
Absorption
Ability to map
Aircraft spectroscopy
methane plumes
remote- using reflected
at the 1–5 m Limited spatial and temporal coverage.
sensing sunlight
scale, direct
measurements or thermal
source attribution.
emissions.
Not as accurate as in situ data.
Absorption
spectroscopy
Global, complete
using reflected
spatial coverage,
sunlight (sensitive
Satellites frequent revisit
to entire Coarse spatial resolution with current instruments (~10
– global time with a
atmospheric km).
coverage single instrument
column) or
(e.g., TROPOMI,
thermal emissions
GOSAT).
(less sensitive to
boundary layer).
Not as accurate as in situ data, emissions not cleanly
resolved.
Limited to sunlit, cloud-free, snow-free scenes.
Absorption
spectroscopy
using reflected
sunlight (sensitive Relatively high
Satellites
to entire resolution (~50 Limited spatial coverage (targeted facilities or
– targeted
atmospheric m x 50 m) (e.g., locations).
monitoring
column) or GHGsat).
thermal emissions
(less sensitive to
boundary layer).
For more information, contact:
The Climate and Clean Air Coalition
Hosted by the United Nations Environment Programme 
1 rue Miollis, 75015 Paris, France
email: secretariat@ccacoalition.org ISBN: 978-92-807-3854-4
Website: www.ccacoalition.org Job No: DTI/2352/PA